venue stringclasses 1
value | title stringlengths 18 162 | abstract stringlengths 252 1.89k | doc_id stringlengths 32 32 | publication_year int64 2.02k 2.02k | sentences listlengths 1 13 | events listlengths 1 24 | document listlengths 50 348 |
|---|---|---|---|---|---|---|---|
ACL | Evaluating Robustness to Input Perturbations for Neural Machine Translation | Neural Machine Translation (NMT) models are sensitive to small perturbations in the input. Robustness to such perturbations is typically measured using translation quality metrics such as BLEU on the noisy input. This paper proposes additional metrics which measure the relative degradation and changes in translation wh... | 680bc4f4a7a98d79fd685b878b72b98e | 2,020 | [
"neural machine translation ( nmt ) models are sensitive to small perturbations in the input .",
"robustness to such perturbations is typically measured using translation quality metrics such as bleu on the noisy input .",
"this paper proposes additional metrics which measure the relative degradation and change... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neural",
"machine",
"translation"
],
"offsets": [
0,
1,
... | [
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"models",
"are",
"sensitive",
"to",
"small",
"perturbations",
"in",
"the",
"input",
".",
"robustness",
"to",
"such",
"perturbations",
"is",
"typically",
"measured",
"using",
"translation",
"quality",
"metrics",
... |
ACL | Regularized Context Gates on Transformer for Machine Translation | Context gates are effective to control the contributions from the source and target contexts in the recurrent neural network (RNN) based neural machine translation (NMT). However, it is challenging to extend them into the advanced Transformer architecture, which is more complicated than RNN. This paper first provides a... | 3f3f1440a9a64fc43bfa2c1038e3dcc6 | 2,020 | [
"context gates are effective to control the contributions from the source and target contexts in the recurrent neural network ( rnn ) based neural machine translation ( nmt ) .",
"however , it is challenging to extend them into the advanced transformer architecture , which is more complicated than rnn .",
"this... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "context gates",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"context",
"gates"
],
"offsets": [
0,
1
]
}
],
"trigger": {
... | [
"context",
"gates",
"are",
"effective",
"to",
"control",
"the",
"contributions",
"from",
"the",
"source",
"and",
"target",
"contexts",
"in",
"the",
"recurrent",
"neural",
"network",
"(",
"rnn",
")",
"based",
"neural",
"machine",
"translation",
"(",
"nmt",
")",... |
ACL | A Joint Neural Model for Information Extraction with Global Features | Most existing joint neural models for Information Extraction (IE) use local task-specific classifiers to predict labels for individual instances (e.g., trigger, relation) regardless of their interactions. For example, a victim of a die event is likely to be a victim of an attack event in the same sentence. In order to ... | 5b6c9a8081c769f1230e51dce95e61fe | 2,020 | [
"most existing joint neural models for information extraction ( ie ) use local task - specific classifiers to predict labels for individual instances ( e . g . , trigger , relation ) regardless of their interactions .",
"for example , a victim of a die event is likely to be a victim of an attack event in the same... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "regardless",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"regardless"
],
"offsets": [
33
]
},
{
"text": "most existing joint neural models",
... | [
"most",
"existing",
"joint",
"neural",
"models",
"for",
"information",
"extraction",
"(",
"ie",
")",
"use",
"local",
"task",
"-",
"specific",
"classifiers",
"to",
"predict",
"labels",
"for",
"individual",
"instances",
"(",
"e",
".",
"g",
".",
",",
"trigger",... |
ACL | Numeracy-600K: Learning Numeracy for Detecting Exaggerated Information in Market Comments | In this paper, we attempt to answer the question of whether neural network models can learn numeracy, which is the ability to predict the magnitude of a numeral at some specific position in a text description. A large benchmark dataset, called Numeracy-600K, is provided for the novel task. We explore several neural net... | faa2be4232018e3714df7ec0db92293c | 2,019 | [
"in this paper , we attempt to answer the question of whether neural network models can learn numeracy , which is the ability to predict the magnitude of a numeral at some specific position in a text description .",
"a large benchmark dataset , called numeracy - 600k , is provided for the novel task .",
"we exp... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "question of whether neural network models can lea... | [
"in",
"this",
"paper",
",",
"we",
"attempt",
"to",
"answer",
"the",
"question",
"of",
"whether",
"neural",
"network",
"models",
"can",
"learn",
"numeracy",
",",
"which",
"is",
"the",
"ability",
"to",
"predict",
"the",
"magnitude",
"of",
"a",
"numeral",
"at... |
ACL | Ordinal and Attribute Aware Response Generation in a Multimodal Dialogue System | Multimodal dialogue systems have opened new frontiers in the traditional goal-oriented dialogue systems. The state-of-the-art dialogue systems are primarily based on unimodal sources, predominantly the text, and hence cannot capture the information present in the other sources such as videos, audios, images etc. With t... | 0473ae79502600f00ff94945ae522282 | 2,019 | [
"multimodal dialogue systems have opened new frontiers in the traditional goal - oriented dialogue systems .",
"the state - of - the - art dialogue systems are primarily based on unimodal sources , predominantly the text , and hence cannot capture the information present in the other sources such as videos , audi... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multimodal dialogue systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"multimodal",
"dialogue",
"systems"
],
"offsets": [
0,
1,
... | [
"multimodal",
"dialogue",
"systems",
"have",
"opened",
"new",
"frontiers",
"in",
"the",
"traditional",
"goal",
"-",
"oriented",
"dialogue",
"systems",
".",
"the",
"state",
"-",
"of",
"-",
"the",
"-",
"art",
"dialogue",
"systems",
"are",
"primarily",
"based",
... |
ACL | What Does BERT Learn about the Structure of Language? | BERT is a recent language representation model that has surprisingly performed well in diverse language understanding benchmarks. This result indicates the possibility that BERT networks capture structural information about language. In this work, we provide novel support for this claim by performing a series of experi... | 2c0e887066fc540afe27bf3aa8f7622c | 2,019 | [
"bert is a recent language representation model that has surprisingly performed well in diverse language understanding benchmarks .",
"this result indicates the possibility that bert networks capture structural information about language .",
"in this work , we provide novel support for this claim by performing ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "bert",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"bert"
],
"offsets": [
0
]
},
{
"text": "in diverse language understanding benchmarks",
... | [
"bert",
"is",
"a",
"recent",
"language",
"representation",
"model",
"that",
"has",
"surprisingly",
"performed",
"well",
"in",
"diverse",
"language",
"understanding",
"benchmarks",
".",
"this",
"result",
"indicates",
"the",
"possibility",
"that",
"bert",
"networks",
... |
ACL | Crowdsourcing and Validating Event-focused Emotion Corpora for German and English | Sentiment analysis has a range of corpora available across multiple languages. For emotion analysis, the situation is more limited, which hinders potential research on crosslingual modeling and the development of predictive models for other languages. In this paper, we fill this gap for German by constructing deISEAR, ... | aae25bab5b001744fcdeb1832793ebde | 2,019 | [
"sentiment analysis has a range of corpora available across multiple languages .",
"for emotion analysis , the situation is more limited , which hinders potential research on crosslingual modeling and the development of predictive models for other languages .",
"in this paper , we fill this gap for german by co... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "sentiment analysis",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"sentiment",
"analysis"
],
"offsets": [
0,
1
]
}
],
"trigger"... | [
"sentiment",
"analysis",
"has",
"a",
"range",
"of",
"corpora",
"available",
"across",
"multiple",
"languages",
".",
"for",
"emotion",
"analysis",
",",
"the",
"situation",
"is",
"more",
"limited",
",",
"which",
"hinders",
"potential",
"research",
"on",
"crossling... |
ACL | Anonymisation Models for Text Data: State of the art, Challenges and Future Directions | This position paper investigates the problem of automated text anonymisation, which is a prerequisite for secure sharing of documents containing sensitive information about individuals. We summarise the key concepts behind text anonymisation and provide a review of current approaches. Anonymisation methods have so far ... | 8a547a45eba5b58a6e70cfec043b2df6 | 2,021 | [
"this position paper investigates the problem of automated text anonymisation , which is a prerequisite for secure sharing of documents containing sensitive information about individuals .",
"we summarise the key concepts behind text anonymisation and provide a review of current approaches .",
"anonymisation me... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "automated text anonymisation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"automated",
"text",
"anonymisation"
],
"offsets": [
7,
8,
... | [
"this",
"position",
"paper",
"investigates",
"the",
"problem",
"of",
"automated",
"text",
"anonymisation",
",",
"which",
"is",
"a",
"prerequisite",
"for",
"secure",
"sharing",
"of",
"documents",
"containing",
"sensitive",
"information",
"about",
"individuals",
".",
... |
ACL | Sources of Transfer in Multilingual Named Entity Recognition | Named-entities are inherently multilingual, and annotations in any given language may be limited. This motivates us to consider polyglot named-entity recognition (NER), where one model is trained using annotated data drawn from more than one language. However, a straightforward implementation of this simple idea does n... | 85923f7b0d0025e0b09e6fb4549ba846 | 2,020 | [
"named - entities are inherently multilingual , and annotations in any given language may be limited .",
"this motivates us to consider polyglot named - entity recognition ( ner ) , where one model is trained using annotated data drawn from more than one language .",
"however , a straightforward implementation ... | [
{
"event_type": "FIN",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Finder",
"tokens": [
"we"
],
"offsets": [
155
]
},
{
"text": "efficiently share",
"nugget_type": "E-FAC",... | [
"named",
"-",
"entities",
"are",
"inherently",
"multilingual",
",",
"and",
"annotations",
"in",
"any",
"given",
"language",
"may",
"be",
"limited",
".",
"this",
"motivates",
"us",
"to",
"consider",
"polyglot",
"named",
"-",
"entity",
"recognition",
"(",
"ner",... |
ACL | Probabilistic Assumptions Matter: Improved Models for Distantly-Supervised Document-Level Question Answering | We address the problem of extractive question answering using document-level distant super-vision, pairing questions and relevant documents with answer strings. We compare previously used probability space and distant supervision assumptions (assumptions on the correspondence between the weak answer string labels and p... | fa4cd2c5a446fa2bcb0e1496ca5b2c77 | 2,020 | [
"we address the problem of extractive question answering using document - level distant super - vision , pairing questions and relevant documents with answer strings .",
"we compare previously used probability space and distant supervision assumptions ( assumptions on the correspondence between the weak answer st... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "questions",
"nugget_type": "FEA",
"argument_type": "BaseComponent",
"tokens": [
"questions"
],
"offsets": [
18
]
},
{
"text": "relevant documents",
"... | [
"we",
"address",
"the",
"problem",
"of",
"extractive",
"question",
"answering",
"using",
"document",
"-",
"level",
"distant",
"super",
"-",
"vision",
",",
"pairing",
"questions",
"and",
"relevant",
"documents",
"with",
"answer",
"strings",
".",
"we",
"compare",
... |
ACL | Tree-Structured Neural Topic Model | This paper presents a tree-structured neural topic model, which has a topic distribution over a tree with an infinite number of branches. Our model parameterizes an unbounded ancestral and fraternal topic distribution by applying doubly-recurrent neural networks. With the help of autoencoding variational Bayes, our mod... | b7b5be32a74da7078cc212b2fb962255 | 2,020 | [
"this paper presents a tree - structured neural topic model , which has a topic distribution over a tree with an infinite number of branches .",
"our model parameterizes an unbounded ancestral and fraternal topic distribution by applying doubly - recurrent neural networks .",
"with the help of autoencoding vari... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "tree - structured neural topic model",
"nugget_type": "APP",
"argument_type": "Content",
"tokens": [
"tree",
"-",
"structured",
"neural",
"topic",
"model"
... | [
"this",
"paper",
"presents",
"a",
"tree",
"-",
"structured",
"neural",
"topic",
"model",
",",
"which",
"has",
"a",
"topic",
"distribution",
"over",
"a",
"tree",
"with",
"an",
"infinite",
"number",
"of",
"branches",
".",
"our",
"model",
"parameterizes",
"an",... |
ACL | A Spreading Activation Framework for Tracking Conceptual Complexity of Texts | We propose an unsupervised approach for assessing conceptual complexity of texts, based on spreading activation. Using DBpedia knowledge graph as a proxy to long-term memory, mentioned concepts become activated and trigger further activation as the text is sequentially traversed. Drawing inspiration from psycholinguist... | f713f820b459a517351c14b060376e4f | 2,019 | [
"we propose an unsupervised approach for assessing conceptual complexity of texts , based on spreading activation .",
"using dbpedia knowledge graph as a proxy to long - term memory , mentioned concepts become activated and trigger further activation as the text is sequentially traversed .",
"drawing inspiratio... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "unsupervised approach",
"nugget_type": "APP... | [
"we",
"propose",
"an",
"unsupervised",
"approach",
"for",
"assessing",
"conceptual",
"complexity",
"of",
"texts",
",",
"based",
"on",
"spreading",
"activation",
".",
"using",
"dbpedia",
"knowledge",
"graph",
"as",
"a",
"proxy",
"to",
"long",
"-",
"term",
"memo... |
ACL | RoMe: A Robust Metric for Evaluating Natural Language Generation | Evaluating Natural Language Generation (NLG) systems is a challenging task. Firstly, the metric should ensure that the generated hypothesis reflects the reference’s semantics. Secondly, it should consider the grammatical quality of the generated sentence. Thirdly, it should be robust enough to handle various surface fo... | 4bf3a83d12fa86b34b055f69d2b06a6b | 2,022 | [
"evaluating natural language generation ( nlg ) systems is a challenging task .",
"firstly , the metric should ensure that the generated hypothesis reflects the reference ’ s semantics .",
"secondly , it should consider the grammatical quality of the generated sentence .",
"thirdly , it should be robust enoug... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "natural language generation ( nlg ) systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"generation",
"(",
"nlg",
")",
... | [
"evaluating",
"natural",
"language",
"generation",
"(",
"nlg",
")",
"systems",
"is",
"a",
"challenging",
"task",
".",
"firstly",
",",
"the",
"metric",
"should",
"ensure",
"that",
"the",
"generated",
"hypothesis",
"reflects",
"the",
"reference",
"’",
"s",
"sema... |
ACL | Self-Supervised Multimodal Opinion Summarization | Recently, opinion summarization, which is the generation of a summary from multiple reviews, has been conducted in a self-supervised manner by considering a sampled review as a pseudo summary. However, non-text data such as image and metadata related to reviews have been considered less often. To use the abundant infor... | 77bbff5bdfb0c184a497cfe8f3d9c28e | 2,021 | [
"recently , opinion summarization , which is the generation of a summary from multiple reviews , has been conducted in a self - supervised manner by considering a sampled review as a pseudo summary .",
"however , non - text data such as image and metadata related to reviews have been considered less often .",
"... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "opinion summarization",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"opinion",
"summarization"
],
"offsets": [
2,
3
]
}
],
"tr... | [
"recently",
",",
"opinion",
"summarization",
",",
"which",
"is",
"the",
"generation",
"of",
"a",
"summary",
"from",
"multiple",
"reviews",
",",
"has",
"been",
"conducted",
"in",
"a",
"self",
"-",
"supervised",
"manner",
"by",
"considering",
"a",
"sampled",
"... |
ACL | Topic Modeling with Wasserstein Autoencoders | We propose a novel neural topic model in the Wasserstein autoencoders (WAE) framework. Unlike existing variational autoencoder based models, we directly enforce Dirichlet prior on the latent document-topic vectors. We exploit the structure of the latent space and apply a suitable kernel in minimizing the Maximum Mean D... | 68226cd75e9367dc0dc264edde0e8b8d | 2,019 | [
"we propose a novel neural topic model in the wasserstein autoencoders ( wae ) framework .",
"unlike existing variational autoencoder based models , we directly enforce dirichlet prior on the latent document - topic vectors .",
"we exploit the structure of the latent space and apply a suitable kernel in minimiz... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "neural topic model",
"nugget_type": "APP",
... | [
"we",
"propose",
"a",
"novel",
"neural",
"topic",
"model",
"in",
"the",
"wasserstein",
"autoencoders",
"(",
"wae",
")",
"framework",
".",
"unlike",
"existing",
"variational",
"autoencoder",
"based",
"models",
",",
"we",
"directly",
"enforce",
"dirichlet",
"prior... |
ACL | Counterfactual Explanations for Natural Language Interfaces | A key challenge facing natural language interfaces is enabling users to understand the capabilities of the underlying system. We propose a novel approach for generating explanations of a natural language interface based on semantic parsing. We focus on counterfactual explanations, which are post-hoc explanations that d... | 9d7738ab2ed015efb4160c9d0b2396b6 | 2,022 | [
"a key challenge facing natural language interfaces is enabling users to understand the capabilities of the underlying system .",
"we propose a novel approach for generating explanations of a natural language interface based on semantic parsing .",
"we focus on counterfactual explanations , which are post - hoc... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "natural language interfaces",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"natural",
"language",
"interfaces"
],
"offsets": [
4,
5,
... | [
"a",
"key",
"challenge",
"facing",
"natural",
"language",
"interfaces",
"is",
"enabling",
"users",
"to",
"understand",
"the",
"capabilities",
"of",
"the",
"underlying",
"system",
".",
"we",
"propose",
"a",
"novel",
"approach",
"for",
"generating",
"explanations",
... |
ACL | The Unstoppable Rise of Computational Linguistics in Deep Learning | In this paper, we trace the history of neural networks applied to natural language understanding tasks, and identify key contributions which the nature of language has made to the development of neural network architectures. We focus on the importance of variable binding and its instantiation in attention-based models,... | cfe907aa3160cea538571f5a223f8654 | 2,020 | [
"in this paper , we trace the history of neural networks applied to natural language understanding tasks , and identify key contributions which the nature of language has made to the development of neural network architectures .",
"we focus on the importance of variable binding and its instantiation in attention ... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "history",
"nugget_type": "TAK",
"... | [
"in",
"this",
"paper",
",",
"we",
"trace",
"the",
"history",
"of",
"neural",
"networks",
"applied",
"to",
"natural",
"language",
"understanding",
"tasks",
",",
"and",
"identify",
"key",
"contributions",
"which",
"the",
"nature",
"of",
"language",
"has",
"made"... |
ACL | Semantic Composition with PSHRG for Derivation Tree Reconstruction from Graph-Based Meaning Representations | We introduce a data-driven approach to generating derivation trees from meaning representation graphs with probabilistic synchronous hyperedge replacement grammar (PSHRG). SHRG has been used to produce meaning representation graphs from texts and syntax trees, but little is known about its viability on the reverse. In ... | c5eba5bf4dca5c52948aacf5437b9d60 | 2,022 | [
"we introduce a data - driven approach to generating derivation trees from meaning representation graphs with probabilistic synchronous hyperedge replacement grammar ( pshrg ) .",
"shrg has been used to produce meaning representation graphs from texts and syntax trees , but little is known about its viability on ... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "data - driven approach",
"nugget_type": "... | [
"we",
"introduce",
"a",
"data",
"-",
"driven",
"approach",
"to",
"generating",
"derivation",
"trees",
"from",
"meaning",
"representation",
"graphs",
"with",
"probabilistic",
"synchronous",
"hyperedge",
"replacement",
"grammar",
"(",
"pshrg",
")",
".",
"shrg",
"has... |
ACL | Updated Headline Generation: Creating Updated Summaries for Evolving News Stories | We propose the task of updated headline generation, in which a system generates a headline for an updated article, considering both the previous article and headline. The system must identify the novel information in the article update, and modify the existing headline accordingly. We create data for this task using th... | 0e8454e35aa61596f5a395077725eb3b | 2,022 | [
"we propose the task of updated headline generation , in which a system generates a headline for an updated article , considering both the previous article and headline .",
"the system must identify the novel information in the article update , and modify the existing headline accordingly .",
"we create data fo... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "task of updated headline generation",
"nugg... | [
"we",
"propose",
"the",
"task",
"of",
"updated",
"headline",
"generation",
",",
"in",
"which",
"a",
"system",
"generates",
"a",
"headline",
"for",
"an",
"updated",
"article",
",",
"considering",
"both",
"the",
"previous",
"article",
"and",
"headline",
".",
"... |
ACL | CipherDAug: Ciphertext based Data Augmentation for Neural Machine Translation | We propose a novel data-augmentation technique for neural machine translation based on ROT-k ciphertexts. ROT-k is a simple letter substitution cipher that replaces a letter in the plaintext with the kth letter after it in the alphabet. We first generate multiple ROT-k ciphertexts using different values of k for the pl... | be118baeb6b7e1c5a79d617dbc9c7f64 | 2,022 | [
"we propose a novel data - augmentation technique for neural machine translation based on rot - k ciphertexts .",
"rot - k is a simple letter substitution cipher that replaces a letter in the plaintext with the kth letter after it in the alphabet .",
"we first generate multiple rot - k ciphertexts using differe... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "data - augmentation technique",
"nugget_typ... | [
"we",
"propose",
"a",
"novel",
"data",
"-",
"augmentation",
"technique",
"for",
"neural",
"machine",
"translation",
"based",
"on",
"rot",
"-",
"k",
"ciphertexts",
".",
"rot",
"-",
"k",
"is",
"a",
"simple",
"letter",
"substitution",
"cipher",
"that",
"replace... |
ACL | Integrating Semantics and Neighborhood Information with Graph-Driven Generative Models for Document Retrieval | With the need of fast retrieval speed and small memory footprint, document hashing has been playing a crucial role in large-scale information retrieval. To generate high-quality hashing code, both semantics and neighborhood information are crucial. However, most existing methods leverage only one of them or simply comb... | 4d5ccc399af5b48424c7856b0991ef9a | 2,021 | [
"with the need of fast retrieval speed and small memory footprint , document hashing has been playing a crucial role in large - scale information retrieval .",
"to generate high - quality hashing code , both semantics and neighborhood information are crucial .",
"however , most existing methods leverage only on... | [
{
"event_type": "MDS",
"arguments": [
{
"text": "semantics and neighborhood information",
"nugget_type": "FEA",
"argument_type": "TriedComponent",
"tokens": [
"semantics",
"and",
"neighborhood",
"information"
],
"o... | [
"with",
"the",
"need",
"of",
"fast",
"retrieval",
"speed",
"and",
"small",
"memory",
"footprint",
",",
"document",
"hashing",
"has",
"been",
"playing",
"a",
"crucial",
"role",
"in",
"large",
"-",
"scale",
"information",
"retrieval",
".",
"to",
"generate",
"h... |
ACL | Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation | Table-to-text generation aims at automatically generating natural text to help people conveniently obtain salient information in tables. Although neural models for table-to-text have achieved remarkable progress, some problems are still overlooked. Previous methods cannot deduce the factual results from the entity’s (p... | 96f3dd4bb70f1852cfba8ed9b7ef2af2 | 2,021 | [
"table - to - text generation aims at automatically generating natural text to help people conveniently obtain salient information in tables .",
"although neural models for table - to - text have achieved remarkable progress , some problems are still overlooked .",
"previous methods cannot deduce the factual re... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "table - to - text generation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"table",
"-",
"to",
"-",
"text",
"generation"
],
"off... | [
"table",
"-",
"to",
"-",
"text",
"generation",
"aims",
"at",
"automatically",
"generating",
"natural",
"text",
"to",
"help",
"people",
"conveniently",
"obtain",
"salient",
"information",
"in",
"tables",
".",
"although",
"neural",
"models",
"for",
"table",
"-",
... |
ACL | Understanding the Language of Political Agreement and Disagreement in Legislative Texts | While national politics often receive the spotlight, the overwhelming majority of legislation proposed, discussed, and enacted is done at the state level. Despite this fact, there is little awareness of the dynamics that lead to adopting these policies. In this paper, we take the first step towards a better understandi... | 518eb043866cc6f3768e99f40cc6b5ad | 2,020 | [
"while national politics often receive the spotlight , the overwhelming majority of legislation proposed , discussed , and enacted is done at the state level .",
"despite this fact , there is little awareness of the dynamics that lead to adopting these policies .",
"in this paper , we take the first step toward... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
48
]
},
{
"text": "using data - driven methods",
"nugget_ty... | [
"while",
"national",
"politics",
"often",
"receive",
"the",
"spotlight",
",",
"the",
"overwhelming",
"majority",
"of",
"legislation",
"proposed",
",",
"discussed",
",",
"and",
"enacted",
"is",
"done",
"at",
"the",
"state",
"level",
".",
"despite",
"this",
"fac... |
ACL | DiS-ReX: A Multilingual Dataset for Distantly Supervised Relation Extraction | Our goal is to study the novel task of distant supervision for multilingual relation extraction (Multi DS-RE). Research in Multi DS-RE has remained limited due to the absence of a reliable benchmarking dataset. The only available dataset for this task, RELX-Distant (Köksal and Özgür, 2020), displays several unrealistic... | 9afb869b32f66211ef8e5a1e403c23fc | 2,022 | [
"our goal is to study the novel task of distant supervision for multilingual relation extraction ( multi ds - re ) .",
"research in multi ds - re has remained limited due to the absence of a reliable benchmarking dataset .",
"the only available dataset for this task , relx - distant ( koksal and ozgur , 2020 ) ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "distant supervision for multilingual relation extraction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"distant",
"supervision",
"for",
"multilingual",
"r... | [
"our",
"goal",
"is",
"to",
"study",
"the",
"novel",
"task",
"of",
"distant",
"supervision",
"for",
"multilingual",
"relation",
"extraction",
"(",
"multi",
"ds",
"-",
"re",
")",
".",
"research",
"in",
"multi",
"ds",
"-",
"re",
"has",
"remained",
"limited",
... |
ACL | M3ED: Multi-modal Multi-scene Multi-label Emotional Dialogue Database | The emotional state of a speaker can be influenced by many different factors in dialogues, such as dialogue scene, dialogue topic, and interlocutor stimulus. The currently available data resources to support such multimodal affective analysis in dialogues are however limited in scale and diversity. In this work, we pro... | 093e78b49b8968e071b83ad9a1d1a05e | 2,022 | [
"the emotional state of a speaker can be influenced by many different factors in dialogues , such as dialogue scene , dialogue topic , and interlocutor stimulus .",
"the currently available data resources to support such multimodal affective analysis in dialogues are however limited in scale and diversity .",
"... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "limited",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"limited"
],
"offsets": [
43
]
},
{
"text": "currently available data resources",
... | [
"the",
"emotional",
"state",
"of",
"a",
"speaker",
"can",
"be",
"influenced",
"by",
"many",
"different",
"factors",
"in",
"dialogues",
",",
"such",
"as",
"dialogue",
"scene",
",",
"dialogue",
"topic",
",",
"and",
"interlocutor",
"stimulus",
".",
"the",
"curr... |
ACL | TaPas: Weakly Supervised Table Parsing via Pre-training | Answering natural language questions over tables is usually seen as a semantic parsing task. To alleviate the collection cost of full logical forms, one popular approach focuses on weak supervision consisting of denotations instead of logical forms. However, training semantic parsers from weak supervision poses difficu... | c2c7725f630241908e83d6840e3e99f0 | 2,020 | [
"answering natural language questions over tables is usually seen as a semantic parsing task .",
"to alleviate the collection cost of full logical forms , one popular approach focuses on weak supervision consisting of denotations instead of logical forms .",
"however , training semantic parsers from weak superv... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "semantic parsing task",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"semantic",
"parsing",
"task"
],
"offsets": [
11,
12,
13
... | [
"answering",
"natural",
"language",
"questions",
"over",
"tables",
"is",
"usually",
"seen",
"as",
"a",
"semantic",
"parsing",
"task",
".",
"to",
"alleviate",
"the",
"collection",
"cost",
"of",
"full",
"logical",
"forms",
",",
"one",
"popular",
"approach",
"foc... |
ACL | Identifying Chinese Opinion Expressions with Extremely-Noisy Crowdsourcing Annotations | Recent works of opinion expression identification (OEI) rely heavily on the quality and scale of the manually-constructed training corpus, which could be extremely difficult to satisfy. Crowdsourcing is one practical solution for this problem, aiming to create a large-scale but quality-unguaranteed corpus. In this work... | 47410d66fdf079756328ab1370a1c036 | 2,022 | [
"recent works of opinion expression identification ( oei ) rely heavily on the quality and scale of the manually - constructed training corpus , which could be extremely difficult to satisfy .",
"crowdsourcing is one practical solution for this problem , aiming to create a large - scale but quality - unguaranteed... | [
{
"event_type": "RWF",
"arguments": [
{
"text": "rely heavily",
"nugget_type": "WEA",
"argument_type": "Fault",
"tokens": [
"rely",
"heavily"
],
"offsets": [
9,
10
]
},
{
"text": "recent... | [
"recent",
"works",
"of",
"opinion",
"expression",
"identification",
"(",
"oei",
")",
"rely",
"heavily",
"on",
"the",
"quality",
"and",
"scale",
"of",
"the",
"manually",
"-",
"constructed",
"training",
"corpus",
",",
"which",
"could",
"be",
"extremely",
"diffic... |
ACL | Hierarchical Transfer Learning for Multi-label Text Classification | Multi-Label Hierarchical Text Classification (MLHTC) is the task of categorizing documents into one or more topics organized in an hierarchical taxonomy. MLHTC can be formulated by combining multiple binary classification problems with an independent classifier for each category. We propose a novel transfer learning ba... | 0e490746b456b563918fbe93d23312cd | 2,019 | [
"multi - label hierarchical text classification ( mlhtc ) is the task of categorizing documents into one or more topics organized in an hierarchical taxonomy .",
"mlhtc can be formulated by combining multiple binary classification problems with an independent classifier for each category .",
"we propose a novel... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multi - label hierarchical text classification",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multi",
"-",
"label",
"hierarchical",
"text",
"cla... | [
"multi",
"-",
"label",
"hierarchical",
"text",
"classification",
"(",
"mlhtc",
")",
"is",
"the",
"task",
"of",
"categorizing",
"documents",
"into",
"one",
"or",
"more",
"topics",
"organized",
"in",
"an",
"hierarchical",
"taxonomy",
".",
"mlhtc",
"can",
"be",
... |
ACL | Argument Pair Extraction via Attention-guided Multi-Layer Multi-Cross Encoding | Argument pair extraction (APE) is a research task for extracting arguments from two passages and identifying potential argument pairs. Prior research work treats this task as a sequence labeling problem and a binary classification problem on two passages that are directly concatenated together, which has a limitation o... | 2f67af64f1be08ae4993378318739579 | 2,021 | [
"argument pair extraction ( ape ) is a research task for extracting arguments from two passages and identifying potential argument pairs .",
"prior research work treats this task as a sequence labeling problem and a binary classification problem on two passages that are directly concatenated together , which has ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "argument pair extraction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"argument",
"pair",
"extraction"
],
"offsets": [
0,
1,
2
... | [
"argument",
"pair",
"extraction",
"(",
"ape",
")",
"is",
"a",
"research",
"task",
"for",
"extracting",
"arguments",
"from",
"two",
"passages",
"and",
"identifying",
"potential",
"argument",
"pairs",
".",
"prior",
"research",
"work",
"treats",
"this",
"task",
"... |
ACL | Multi-source Meta Transfer for Low Resource Multiple-Choice Question Answering | Multiple-choice question answering (MCQA) is one of the most challenging tasks in machine reading comprehension since it requires more advanced reading comprehension skills such as logical reasoning, summarization, and arithmetic operations. Unfortunately, most existing MCQA datasets are small in size, which increases ... | 184fd60faf99b39c01a6a0e45489c5d3 | 2,020 | [
"multiple - choice question answering ( mcqa ) is one of the most challenging tasks in machine reading comprehension since it requires more advanced reading comprehension skills such as logical reasoning , summarization , and arithmetic operations .",
"unfortunately , most existing mcqa datasets are small in size... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multiple - choice question answering",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multiple",
"-",
"choice",
"question",
"answering"
],
"... | [
"multiple",
"-",
"choice",
"question",
"answering",
"(",
"mcqa",
")",
"is",
"one",
"of",
"the",
"most",
"challenging",
"tasks",
"in",
"machine",
"reading",
"comprehension",
"since",
"it",
"requires",
"more",
"advanced",
"reading",
"comprehension",
"skills",
"suc... |
ACL | Unsupervised Bilingual Word Embedding Agreement for Unsupervised Neural Machine Translation | Unsupervised bilingual word embedding (UBWE), together with other technologies such as back-translation and denoising, has helped unsupervised neural machine translation (UNMT) achieve remarkable results in several language pairs. In previous methods, UBWE is first trained using non-parallel monolingual corpora and the... | f72ea4294064ef2d5fb8982d648e6716 | 2,019 | [
"unsupervised bilingual word embedding ( ubwe ) , together with other technologies such as back - translation and denoising , has helped unsupervised neural machine translation ( unmt ) achieve remarkable results in several language pairs .",
"in previous methods , ubwe is first trained using non - parallel monol... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "unsupervised neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"unsupervised",
"neural",
"machine",
"translation"
],
"offsets... | [
"unsupervised",
"bilingual",
"word",
"embedding",
"(",
"ubwe",
")",
",",
"together",
"with",
"other",
"technologies",
"such",
"as",
"back",
"-",
"translation",
"and",
"denoising",
",",
"has",
"helped",
"unsupervised",
"neural",
"machine",
"translation",
"(",
"un... |
ACL | Generating Sentences from Disentangled Syntactic and Semantic Spaces | Variational auto-encoders (VAEs) are widely used in natural language generation due to the regularization of the latent space. However, generating sentences from the continuous latent space does not explicitly model the syntactic information. In this paper, we propose to generate sentences from disentangled syntactic a... | 4ccf9178462d0328d9817688197a3a15 | 2,019 | [
"variational auto - encoders ( vaes ) are widely used in natural language generation due to the regularization of the latent space .",
"however , generating sentences from the continuous latent space does not explicitly model the syntactic information .",
"in this paper , we propose to generate sentences from d... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "variational auto - encoders",
"nugget_type": "MOD",
"argument_type": "Target",
"tokens": [
"variational",
"auto",
"-",
"encoders"
],
"offsets": [
0,
... | [
"variational",
"auto",
"-",
"encoders",
"(",
"vaes",
")",
"are",
"widely",
"used",
"in",
"natural",
"language",
"generation",
"due",
"to",
"the",
"regularization",
"of",
"the",
"latent",
"space",
".",
"however",
",",
"generating",
"sentences",
"from",
"the",
... |
ACL | A Simple Recipe for Multilingual Grammatical Error Correction | This paper presents a simple recipe to trainstate-of-the-art multilingual Grammatical Error Correction (GEC) models. We achieve this by first proposing a language-agnostic method to generate a large number of synthetic examples. The second ingredient is to use large-scale multilingual language models (up to 11B paramet... | 569048fdf8854be616c481127e2904f2 | 2,021 | [
"this paper presents a simple recipe to trainstate - of - the - art multilingual grammatical error correction ( gec ) models .",
"we achieve this by first proposing a language - agnostic method to generate a large number of synthetic examples .",
"the second ingredient is to use large - scale multilingual langu... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multilingual grammatical error correction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multilingual",
"grammatical",
"error",
"correction"
],
"off... | [
"this",
"paper",
"presents",
"a",
"simple",
"recipe",
"to",
"trainstate",
"-",
"of",
"-",
"the",
"-",
"art",
"multilingual",
"grammatical",
"error",
"correction",
"(",
"gec",
")",
"models",
".",
"we",
"achieve",
"this",
"by",
"first",
"proposing",
"a",
"la... |
ACL | Detecting Concealed Information in Text and Speech | Motivated by infamous cheating scandals in the media industry, the wine industry, and political campaigns, we address the problem of detecting concealed information in technical settings. In this work, we explore acoustic-prosodic and linguistic indicators of information concealment by collecting a unique corpus of pro... | 491fa4ec3fe22ce1d6599807af26623b | 2,019 | [
"motivated by infamous cheating scandals in the media industry , the wine industry , and political campaigns , we address the problem of detecting concealed information in technical settings .",
"in this work , we explore acoustic - prosodic and linguistic indicators of information concealment by collecting a uni... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
18
]
},
{
"text": "problem of detecting concealed information in te... | [
"motivated",
"by",
"infamous",
"cheating",
"scandals",
"in",
"the",
"media",
"industry",
",",
"the",
"wine",
"industry",
",",
"and",
"political",
"campaigns",
",",
"we",
"address",
"the",
"problem",
"of",
"detecting",
"concealed",
"information",
"in",
"technical... |
ACL | Integrating Vectorized Lexical Constraints for Neural Machine Translation | Lexically constrained neural machine translation (NMT), which controls the generation of NMT models with pre-specified constraints, is important in many practical scenarios. Due to the representation gap between discrete constraints and continuous vectors in NMT models, most existing works choose to construct synthetic... | 0fec61a3e5d192e9c2a08d09f3b4a845 | 2,022 | [
"lexically constrained neural machine translation ( nmt ) , which controls the generation of nmt models with pre - specified constraints , is important in many practical scenarios .",
"due to the representation gap between discrete constraints and continuous vectors in nmt models , most existing works choose to c... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "lexically constrained neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"lexically",
"constrained",
"neural",
"machine",
"translati... | [
"lexically",
"constrained",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
",",
"which",
"controls",
"the",
"generation",
"of",
"nmt",
"models",
"with",
"pre",
"-",
"specified",
"constraints",
",",
"is",
"important",
"in",
"many",
"practical",
"scenarios",... |
ACL | DYPLOC: Dynamic Planning of Content Using Mixed Language Models for Text Generation | We study the task of long-form opinion text generation, which faces at least two distinct challenges. First, existing neural generation models fall short of coherence, thus requiring efficient content planning. Second, diverse types of information are needed to guide the generator to cover both subjective and objective... | bb7266e4ed4dfd382f5c9898e8cda4af | 2,021 | [
"we study the task of long - form opinion text generation , which faces at least two distinct challenges .",
"first , existing neural generation models fall short of coherence , thus requiring efficient content planning .",
"second , diverse types of information are needed to guide the generator to cover both s... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "task of long - form opinion text generation",
... | [
"we",
"study",
"the",
"task",
"of",
"long",
"-",
"form",
"opinion",
"text",
"generation",
",",
"which",
"faces",
"at",
"least",
"two",
"distinct",
"challenges",
".",
"first",
",",
"existing",
"neural",
"generation",
"models",
"fall",
"short",
"of",
"coherenc... |
ACL | Syntax-Infused Variational Autoencoder for Text Generation | We present a syntax-infused variational autoencoder (SIVAE), that integrates sentences with their syntactic trees to improve the grammar of generated sentences. Distinct from existing VAE-based text generative models, SIVAE contains two separate latent spaces, for sentences and syntactic trees. The evidence lower bound... | e30dd814893d3c06bc54506b14259eeb | 2,019 | [
"we present a syntax - infused variational autoencoder ( sivae ) , that integrates sentences with their syntactic trees to improve the grammar of generated sentences .",
"distinct from existing vae - based text generative models , sivae contains two separate latent spaces , for sentences and syntactic trees .",
... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "syntax - infused variational autoencoder",
... | [
"we",
"present",
"a",
"syntax",
"-",
"infused",
"variational",
"autoencoder",
"(",
"sivae",
")",
",",
"that",
"integrates",
"sentences",
"with",
"their",
"syntactic",
"trees",
"to",
"improve",
"the",
"grammar",
"of",
"generated",
"sentences",
".",
"distinct",
... |
ACL | A Targeted Assessment of Incremental Processing in Neural Language Models and Humans | We present a targeted, scaled-up comparison of incremental processing in humans and neural language models by collecting by-word reaction time data for sixteen different syntactic test suites across a range of structural phenomena. Human reaction time data comes from a novel online experimental paradigm called the Inte... | 5ab8f1e1d13ed475e5fca0f682607efa | 2,021 | [
"we present a targeted , scaled - up comparison of incremental processing in humans and neural language models by collecting by - word reaction time data for sixteen different syntactic test suites across a range of structural phenomena .",
"human reaction time data comes from a novel online experimental paradigm... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "targeted , scaled - up comparison of incremental pr... | [
"we",
"present",
"a",
"targeted",
",",
"scaled",
"-",
"up",
"comparison",
"of",
"incremental",
"processing",
"in",
"humans",
"and",
"neural",
"language",
"models",
"by",
"collecting",
"by",
"-",
"word",
"reaction",
"time",
"data",
"for",
"sixteen",
"different"... |
ACL | Conversations Are Not Flat: Modeling the Dynamic Information Flow across Dialogue Utterances | Nowadays, open-domain dialogue models can generate acceptable responses according to the historical context based on the large-scale pre-trained language models. However, they generally concatenate the dialogue history directly as the model input to predict the response, which we named as the flat pattern and ignores t... | c5cb5911515ea16b0e78ad9bed6cab75 | 2,021 | [
"nowadays , open - domain dialogue models can generate acceptable responses according to the historical context based on the large - scale pre - trained language models .",
"however , they generally concatenate the dialogue history directly as the model input to predict the response , which we named as the flat p... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "open - domain dialogue models",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"open",
"-",
"domain",
"dialogue",
"models"
],
"offsets": [
... | [
"nowadays",
",",
"open",
"-",
"domain",
"dialogue",
"models",
"can",
"generate",
"acceptable",
"responses",
"according",
"to",
"the",
"historical",
"context",
"based",
"on",
"the",
"large",
"-",
"scale",
"pre",
"-",
"trained",
"language",
"models",
".",
"howev... |
ACL | Dialogue State Tracking with Explicit Slot Connection Modeling | Recent proposed approaches have made promising progress in dialogue state tracking (DST). However, in multi-domain scenarios, ellipsis and reference are frequently adopted by users to express values that have been mentioned by slots from other domains. To handle these phenomena, we propose a Dialogue State Tracking wit... | 84c72c0671465b6434f2492e51a7ff4b | 2,020 | [
"recent proposed approaches have made promising progress in dialogue state tracking ( dst ) .",
"however , in multi - domain scenarios , ellipsis and reference are frequently adopted by users to express values that have been mentioned by slots from other domains .",
"to handle these phenomena , we propose a dia... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
49
]
},
{
"text": "dialogue state tracking with slot connections mode... | [
"recent",
"proposed",
"approaches",
"have",
"made",
"promising",
"progress",
"in",
"dialogue",
"state",
"tracking",
"(",
"dst",
")",
".",
"however",
",",
"in",
"multi",
"-",
"domain",
"scenarios",
",",
"ellipsis",
"and",
"reference",
"are",
"frequently",
"adop... |
ACL | Multimodal Multi-Speaker Merger & Acquisition Financial Modeling: A New Task, Dataset, and Neural Baselines | Risk prediction is an essential task in financial markets. Merger and Acquisition (M&A) calls provide key insights into the claims made by company executives about the restructuring of the financial firms. Extracting vocal and textual cues from M&A calls can help model the risk associated with such financial activities... | 03bf08b7f0cfaf1b4a6d04fb8fe80374 | 2,021 | [
"risk prediction is an essential task in financial markets .",
"merger and acquisition ( m & a ) calls provide key insights into the claims made by company executives about the restructuring of the financial firms .",
"extracting vocal and textual cues from m & a calls can help model the risk associated with su... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "risk prediction",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"risk",
"prediction"
],
"offsets": [
0,
1
]
}
],
"trigger": {
... | [
"risk",
"prediction",
"is",
"an",
"essential",
"task",
"in",
"financial",
"markets",
".",
"merger",
"and",
"acquisition",
"(",
"m",
"&",
"a",
")",
"calls",
"provide",
"key",
"insights",
"into",
"the",
"claims",
"made",
"by",
"company",
"executives",
"about",... |
ACL | Exploring the Efficacy of Automatically Generated Counterfactuals for Sentiment Analysis | While state-of-the-art NLP models have been achieving the excellent performance of a wide range of tasks in recent years, important questions are being raised about their robustness and their underlying sensitivity to systematic biases that may exist in their training and test data. Such issues come to be manifest in p... | 8aaad5ce9eaedf86fb76985fb6d6f7b7 | 2,021 | [
"while state - of - the - art nlp models have been achieving the excellent performance of a wide range of tasks in recent years , important questions are being raised about their robustness and their underlying sensitivity to systematic biases that may exist in their training and test data .",
"such issues come t... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "one recent solution",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"one",
"recent",
"solution"
],
"offsets": [
73,
74,
75
... | [
"while",
"state",
"-",
"of",
"-",
"the",
"-",
"art",
"nlp",
"models",
"have",
"been",
"achieving",
"the",
"excellent",
"performance",
"of",
"a",
"wide",
"range",
"of",
"tasks",
"in",
"recent",
"years",
",",
"important",
"questions",
"are",
"being",
"raised... |
ACL | Tagged Back-translation Revisited: Why Does It Really Work? | In this paper, we show that neural machine translation (NMT) systems trained on large back-translated data overfit some of the characteristics of machine-translated texts. Such NMT systems better translate human-produced translations, i.e., translationese, but may largely worsen the translation quality of original text... | c675bd09a84675213cc252c5dd2dfe83 | 2,020 | [
"in this paper , we show that neural machine translation ( nmt ) systems trained on large back - translated data overfit some of the characteristics of machine - translated texts .",
"such nmt systems better translate human - produced translations , i . e . , translationese , but may largely worsen the translatio... | [
{
"event_type": "FIN",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Finder",
"tokens": [
"we"
],
"offsets": [
4
]
},
{
"text": "overfit",
"nugget_type": "E-FAC",
"ar... | [
"in",
"this",
"paper",
",",
"we",
"show",
"that",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"systems",
"trained",
"on",
"large",
"back",
"-",
"translated",
"data",
"overfit",
"some",
"of",
"the",
"characteristics",
"of",
"machine",
"-",
"translate... |
ACL | Learned Incremental Representations for Parsing | We present an incremental syntactic representation that consists of assigning a single discrete label to each word in a sentence, where the label is predicted using strictly incremental processing of a prefix of the sentence, and the sequence of labels for a sentence fully determines a parse tree. Our goal is to induce... | d3b71dc86df07334ad47020f6b3fdbee | 2,022 | [
"we present an incremental syntactic representation that consists of assigning a single discrete label to each word in a sentence , where the label is predicted using strictly incremental processing of a prefix of the sentence , and the sequence of labels for a sentence fully determines a parse tree .",
"our goal... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "incremental syntactic representation",
"nug... | [
"we",
"present",
"an",
"incremental",
"syntactic",
"representation",
"that",
"consists",
"of",
"assigning",
"a",
"single",
"discrete",
"label",
"to",
"each",
"word",
"in",
"a",
"sentence",
",",
"where",
"the",
"label",
"is",
"predicted",
"using",
"strictly",
"... |
ACL | Evaluating Entity Disambiguation and the Role of Popularity in Retrieval-Based NLP | Retrieval is a core component for open-domain NLP tasks. In open-domain tasks, multiple entities can share a name, making disambiguation an inherent yet under-explored problem. We propose an evaluation benchmark for assessing the entity disambiguation capabilities of these retrievers, which we call Ambiguous Entity Ret... | c73b3ff3c8a5931212a2fb07cc8276d0 | 2,021 | [
"retrieval is a core component for open - domain nlp tasks .",
"in open - domain tasks , multiple entities can share a name , making disambiguation an inherent yet under - explored problem .",
"we propose an evaluation benchmark for assessing the entity disambiguation capabilities of these retrievers , which we... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "disambiguation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"disambiguation"
],
"offsets": [
26
]
}
],
"trigger": {
"text": "making",
... | [
"retrieval",
"is",
"a",
"core",
"component",
"for",
"open",
"-",
"domain",
"nlp",
"tasks",
".",
"in",
"open",
"-",
"domain",
"tasks",
",",
"multiple",
"entities",
"can",
"share",
"a",
"name",
",",
"making",
"disambiguation",
"an",
"inherent",
"yet",
"under... |
ACL | BIGPATENT: A Large-Scale Dataset for Abstractive and Coherent Summarization | Most existing text summarization datasets are compiled from the news domain, where summaries have a flattened discourse structure. In such datasets, summary-worthy content often appears in the beginning of input articles. Moreover, large segments from input articles are present verbatim in their respective summaries. T... | 039d60d5fae924629888f9159a474756 | 2,019 | [
"most existing text summarization datasets are compiled from the news domain , where summaries have a flattened discourse structure .",
"in such datasets , summary - worthy content often appears in the beginning of input articles .",
"moreover , large segments from input articles are present verbatim in their r... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "text summarization datasets",
"nugget_type": "DST",
"argument_type": "Target",
"tokens": [
"text",
"summarization",
"datasets"
],
"offsets": [
2,
3,
... | [
"most",
"existing",
"text",
"summarization",
"datasets",
"are",
"compiled",
"from",
"the",
"news",
"domain",
",",
"where",
"summaries",
"have",
"a",
"flattened",
"discourse",
"structure",
".",
"in",
"such",
"datasets",
",",
"summary",
"-",
"worthy",
"content",
... |
ACL | Examining the Inductive Bias of Neural Language Models with Artificial Languages | Since language models are used to model a wide variety of languages, it is natural to ask whether the neural architectures used for the task have inductive biases towards modeling particular types of languages. Investigation of these biases has proved complicated due to the many variables that appear in the experimenta... | ffaf12ef9d8be33e1c606ebe0d8c363c | 2,021 | [
"since language models are used to model a wide variety of languages , it is natural to ask whether the neural architectures used for the task have inductive biases towards modeling particular types of languages .",
"investigation of these biases has proved complicated due to the many variables that appear in the... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "inductive biases",
"nugget_type": "FEA",
"argument_type": "Target",
"tokens": [
"inductive",
"biases"
],
"offsets": [
27,
28
]
}
],
"trigger": ... | [
"since",
"language",
"models",
"are",
"used",
"to",
"model",
"a",
"wide",
"variety",
"of",
"languages",
",",
"it",
"is",
"natural",
"to",
"ask",
"whether",
"the",
"neural",
"architectures",
"used",
"for",
"the",
"task",
"have",
"inductive",
"biases",
"toward... |
ACL | Improving Zero-Shot Translation by Disentangling Positional Information | Multilingual neural machine translation has shown the capability of directly translating between language pairs unseen in training, i.e. zero-shot translation. Despite being conceptually attractive, it often suffers from low output quality. The difficulty of generalizing to new translation directions suggests the model... | e797b967628fad130a9bb3e51e061909 | 2,021 | [
"multilingual neural machine translation has shown the capability of directly translating between language pairs unseen in training , i . e . zero - shot translation .",
"despite being conceptually attractive , it often suffers from low output quality .",
"the difficulty of generalizing to new translation direc... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "multilingual neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"multilingual",
"neural",
"machine",
"translation"
],
"offsets... | [
"multilingual",
"neural",
"machine",
"translation",
"has",
"shown",
"the",
"capability",
"of",
"directly",
"translating",
"between",
"language",
"pairs",
"unseen",
"in",
"training",
",",
"i",
".",
"e",
".",
"zero",
"-",
"shot",
"translation",
".",
"despite",
"... |
ACL | SRL4E – Semantic Role Labeling for Emotions: A Unified Evaluation Framework | In the field of sentiment analysis, several studies have highlighted that a single sentence may express multiple, sometimes contrasting, sentiments and emotions, each with its own experiencer, target and/or cause. To this end, over the past few years researchers have started to collect and annotate data manually, in or... | 377f5cd7515f630117d30220877f5dd4 | 2,022 | [
"in the field of sentiment analysis , several studies have highlighted that a single sentence may express multiple , sometimes contrasting , sentiments and emotions , each with its own experiencer , target and / or cause .",
"to this end , over the past few years researchers have started to collect and annotate d... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "sentiment analysis",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"sentiment",
"analysis"
],
"offsets": [
4,
5
]
}
],
"trigger"... | [
"in",
"the",
"field",
"of",
"sentiment",
"analysis",
",",
"several",
"studies",
"have",
"highlighted",
"that",
"a",
"single",
"sentence",
"may",
"express",
"multiple",
",",
"sometimes",
"contrasting",
",",
"sentiments",
"and",
"emotions",
",",
"each",
"with",
... |
ACL | Beyond Possession Existence: Duration and Co-Possession | This paper introduces two tasks: determining (a) the duration of possession relations and (b) co-possessions, i.e., whether multiple possessors possess a possessee at the same time. We present new annotations on top of corpora annotating possession existence and experimental results. Regarding possession duration, we d... | 11adb75b9d29c1e45959df435a7f3dbb | 2,020 | [
"this paper introduces two tasks : determining ( a ) the duration of possession relations and ( b ) co - possessions , i . e . , whether multiple possessors possess a possessee at the same time .",
"we present new annotations on top of corpora annotating possession existence and experimental results .",
"regard... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "duration of possession relations",
"nugget_type": "FEA",
"argument_type": "Content",
"tokens": [
"duration",
"of",
"possession",
"relations"
],
"offsets": [
... | [
"this",
"paper",
"introduces",
"two",
"tasks",
":",
"determining",
"(",
"a",
")",
"the",
"duration",
"of",
"possession",
"relations",
"and",
"(",
"b",
")",
"co",
"-",
"possessions",
",",
"i",
".",
"e",
".",
",",
"whether",
"multiple",
"possessors",
"poss... |
ACL | Human vs. Muppet: A Conservative Estimate of Human Performance on the GLUE Benchmark | The GLUE benchmark (Wang et al., 2019b) is a suite of language understanding tasks which has seen dramatic progress in the past year, with average performance moving from 70.0 at launch to 83.9, state of the art at the time of writing (May 24, 2019). Here, we measure human performance on the benchmark, in order to lear... | c78d6c3c14e25ad2551cbffdcdd1ca3f | 2,019 | [
"the glue benchmark ( wang et al . , 2019b ) is a suite of language understanding tasks which has seen dramatic progress in the past year , with average performance moving from 70 . 0 at launch to 83 . 9 , state of the art at the time of writing ( may 24 , 2019 ) .",
"here , we measure human performance on the be... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
61
]
},
{
"text": "human performance",
"nugget_type": "TAK"... | [
"the",
"glue",
"benchmark",
"(",
"wang",
"et",
"al",
".",
",",
"2019b",
")",
"is",
"a",
"suite",
"of",
"language",
"understanding",
"tasks",
"which",
"has",
"seen",
"dramatic",
"progress",
"in",
"the",
"past",
"year",
",",
"with",
"average",
"performance",... |
ACL | Conversational Word Embedding for Retrieval-Based Dialog System | Human conversations contain many types of information, e.g., knowledge, common sense, and language habits. In this paper, we propose a conversational word embedding method named PR-Embedding, which utilizes the conversation pairs <post, reply> to learn word embedding. Different from previous works, PR-Embedding uses th... | 5b0def0cf7b8a40282b48a6200d08f20 | 2,020 | [
"human conversations contain many types of information , e . g . , knowledge , common sense , and language habits .",
"in this paper , we propose a conversational word embedding method named pr - embedding , which utilizes the conversation pairs < post , reply > to learn word embedding .",
"different from previ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "human conversations",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"human",
"conversations"
],
"offsets": [
0,
1
]
}
],
"trigge... | [
"human",
"conversations",
"contain",
"many",
"types",
"of",
"information",
",",
"e",
".",
"g",
".",
",",
"knowledge",
",",
"common",
"sense",
",",
"and",
"language",
"habits",
".",
"in",
"this",
"paper",
",",
"we",
"propose",
"a",
"conversational",
"word",... |
ACL | Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection | The ability to control for the kinds of information encoded in neural representation has a variety of use cases, especially in light of the challenge of interpreting these models. We present Iterative Null-space Projection (INLP), a novel method for removing information from neural representations. Our method is based ... | 653c684e8c4c696773acfff884d5ab43 | 2,020 | [
"the ability to control for the kinds of information encoded in neural representation has a variety of use cases , especially in light of the challenge of interpreting these models .",
"we present iterative null - space projection ( inlp ) , a novel method for removing information from neural representations .",
... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "information encoded in neural representation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"information",
"encoded",
"in",
"neural",
"representation"
... | [
"the",
"ability",
"to",
"control",
"for",
"the",
"kinds",
"of",
"information",
"encoded",
"in",
"neural",
"representation",
"has",
"a",
"variety",
"of",
"use",
"cases",
",",
"especially",
"in",
"light",
"of",
"the",
"challenge",
"of",
"interpreting",
"these",
... |
ACL | Multilingual Pre-training with Language and Task Adaptation for Multilingual Text Style Transfer | We exploit the pre-trained seq2seq model mBART for multilingual text style transfer. Using machine translated data as well as gold aligned English sentences yields state-of-the-art results in the three target languages we consider. Besides, in view of the general scarcity of parallel data, we propose a modular approach... | cd9b1e8dcad39f88d885fc2e16488894 | 2,022 | [
"we exploit the pre - trained seq2seq model mbart for multilingual text style transfer .",
"using machine translated data as well as gold aligned english sentences yields state - of - the - art results in the three target languages we consider .",
"besides , in view of the general scarcity of parallel data , we... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "multilingual text style transfer",
"nugge... | [
"we",
"exploit",
"the",
"pre",
"-",
"trained",
"seq2seq",
"model",
"mbart",
"for",
"multilingual",
"text",
"style",
"transfer",
".",
"using",
"machine",
"translated",
"data",
"as",
"well",
"as",
"gold",
"aligned",
"english",
"sentences",
"yields",
"state",
"-"... |
ACL | Syntax-Aware Opinion Role Labeling with Dependency Graph Convolutional Networks | Opinion role labeling (ORL) is a fine-grained opinion analysis task and aims to answer “who expressed what kind of sentiment towards what?”. Due to the scarcity of labeled data, ORL remains challenging for data-driven methods. In this work, we try to enhance neural ORL models with syntactic knowledge by comparing and i... | 9c698d0250e04493d021cfeea9d5127a | 2,020 | [
"opinion role labeling ( orl ) is a fine - grained opinion analysis task and aims to answer “ who expressed what kind of sentiment towards what ? ” .",
"due to the scarcity of labeled data , orl remains challenging for data - driven methods .",
"in this work , we try to enhance neural orl models with syntactic ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "orl",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"orl"
],
"offsets": [
139
]
}
],
"trigger": {
"text": "task",
"tokens": [
"t... | [
"opinion",
"role",
"labeling",
"(",
"orl",
")",
"is",
"a",
"fine",
"-",
"grained",
"opinion",
"analysis",
"task",
"and",
"aims",
"to",
"answer",
"“",
"who",
"expressed",
"what",
"kind",
"of",
"sentiment",
"towards",
"what",
"?",
"”",
".",
"due",
"to",
... |
ACL | Human Evaluation and Correlation with Automatic Metrics in Consultation Note Generation | In recent years, machine learning models have rapidly become better at generating clinical consultation notes; yet, there is little work on how to properly evaluate the generated consultation notes to understand the impact they may have on both the clinician using them and the patient’s clinical safety.To address this ... | 248c1076aab69b3eb055a4ebe37b8595 | 2,022 | [
"in recent years , machine learning models have rapidly become better at generating clinical consultation notes ; yet , there is little work on how to properly evaluate the generated consultation notes to understand the impact they may have on both the clinician using them and the patient ’ s clinical safety .",
... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "clinical consultation notes",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"clinical",
"consultation",
"notes"
],
"offsets": [
13,
14,
... | [
"in",
"recent",
"years",
",",
"machine",
"learning",
"models",
"have",
"rapidly",
"become",
"better",
"at",
"generating",
"clinical",
"consultation",
"notes",
";",
"yet",
",",
"there",
"is",
"little",
"work",
"on",
"how",
"to",
"properly",
"evaluate",
"the",
... |
ACL | Better Language Model with Hypernym Class Prediction | Class-based language models (LMs) have been long devised to address context sparsity in n-gram LMs. In this study, we revisit this approach in the context of neural LMs. We hypothesize that class-based prediction leads to an implicit context aggregation for similar words and thus can improve generalization for rare wor... | a1ac6bf5a5180752e7eacf550d94034f | 2,022 | [
"class - based language models ( lms ) have been long devised to address context sparsity in n - gram lms .",
"in this study , we revisit this approach in the context of neural lms .",
"we hypothesize that class - based prediction leads to an implicit context aggregation for similar words and thus can improve g... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "class - based language models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"class",
"-",
"based",
"language",
"models"
],
"offsets": [
... | [
"class",
"-",
"based",
"language",
"models",
"(",
"lms",
")",
"have",
"been",
"long",
"devised",
"to",
"address",
"context",
"sparsity",
"in",
"n",
"-",
"gram",
"lms",
".",
"in",
"this",
"study",
",",
"we",
"revisit",
"this",
"approach",
"in",
"the",
"... |
ACL | Dense Procedure Captioning in Narrated Instructional Videos | Understanding narrated instructional videos is important for both research and real-world web applications. Motivated by video dense captioning, we propose a model to generate procedure captions from narrated instructional videos which are a sequence of step-wise clips with description. Previous works on video dense ca... | c945d3083053612bcbe698a011f73bc5 | 2,019 | [
"understanding narrated instructional videos is important for both research and real - world web applications .",
"motivated by video dense captioning , we propose a model to generate procedure captions from narrated instructional videos which are a sequence of step - wise clips with description .",
"previous w... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
22
]
},
{
"text": "model",
"nugget_type": "APP",
"arg... | [
"understanding",
"narrated",
"instructional",
"videos",
"is",
"important",
"for",
"both",
"research",
"and",
"real",
"-",
"world",
"web",
"applications",
".",
"motivated",
"by",
"video",
"dense",
"captioning",
",",
"we",
"propose",
"a",
"model",
"to",
"generate"... |
ACL | Energy-Based Reranking: Improving Neural Machine Translation Using Energy-Based Models | The discrepancy between maximum likelihood estimation (MLE) and task measures such as BLEU score has been studied before for autoregressive neural machine translation (NMT) and resulted in alternative training algorithms (Ranzato et al., 2016; Norouzi et al., 2016; Shen et al., 2016; Wu et al., 2018). However, MLE trai... | c3fa40dd97926cd60c3388960ad315fc | 2,021 | [
"the discrepancy between maximum likelihood estimation ( mle ) and task measures such as bleu score has been studied before for autoregressive neural machine translation ( nmt ) and resulted in alternative training algorithms ( ranzato et al . , 2016 ; norouzi et al . , 2016 ; shen et al . , 2016 ; wu et al . , 201... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "maximum likelihood estimation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"maximum",
"likelihood",
"estimation"
],
"offsets": [
3,
4,
... | [
"the",
"discrepancy",
"between",
"maximum",
"likelihood",
"estimation",
"(",
"mle",
")",
"and",
"task",
"measures",
"such",
"as",
"bleu",
"score",
"has",
"been",
"studied",
"before",
"for",
"autoregressive",
"neural",
"machine",
"translation",
"(",
"nmt",
")",
... |
ACL | DYLE: Dynamic Latent Extraction for Abstractive Long-Input Summarization | Transformer-based models have achieved state-of-the-art performance on short-input summarization. However, they still struggle with summarizing longer text. In this paper, we present DYLE, a novel dynamic latent extraction approach for abstractive long-input summarization. DYLE jointly trains an extractor and a generat... | c31baff9c943974a5f422d1d0600ffae | 2,022 | [
"transformer - based models have achieved state - of - the - art performance on short - input summarization .",
"however , they still struggle with summarizing longer text .",
"in this paper , we present dyle , a novel dynamic latent extraction approach for abstractive long - input summarization .",
"dyle joi... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "short - input summarization",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"short",
"-",
"input",
"summarization"
],
"offsets": [
15,
... | [
"transformer",
"-",
"based",
"models",
"have",
"achieved",
"state",
"-",
"of",
"-",
"the",
"-",
"art",
"performance",
"on",
"short",
"-",
"input",
"summarization",
".",
"however",
",",
"they",
"still",
"struggle",
"with",
"summarizing",
"longer",
"text",
"."... |
ACL | XDBERT: Distilling Visual Information to BERT from Cross-Modal Systems to Improve Language Understanding | Transformer-based models are widely used in natural language understanding (NLU) tasks, and multimodal transformers have been effective in visual-language tasks. This study explores distilling visual information from pretrained multimodal transformers to pretrained language encoders. Our framework is inspired by cross-... | 3b81a6f9f9cacbec676d6873a98c0666 | 2,022 | [
"transformer - based models are widely used in natural language understanding ( nlu ) tasks , and multimodal transformers have been effective in visual - language tasks .",
"this study explores distilling visual information from pretrained multimodal transformers to pretrained language encoders .",
"our framewo... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "transformer - based models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"transformer",
"-",
"based",
"models"
],
"offsets": [
0,
... | [
"transformer",
"-",
"based",
"models",
"are",
"widely",
"used",
"in",
"natural",
"language",
"understanding",
"(",
"nlu",
")",
"tasks",
",",
"and",
"multimodal",
"transformers",
"have",
"been",
"effective",
"in",
"visual",
"-",
"language",
"tasks",
".",
"this"... |
ACL | Guided Attention Multimodal Multitask Financial Forecasting with Inter-Company Relationships and Global and Local News | Most works on financial forecasting use information directly associated with individual companies (e.g., stock prices, news on the company) to predict stock returns for trading. We refer to such company-specific information as local information. Stock returns may also be influenced by global information (e.g., news on ... | 387f7a4111a51dd84b2c82866a1ef7a5 | 2,022 | [
"most works on financial forecasting use information directly associated with individual companies ( e . g . , stock prices , news on the company ) to predict stock returns for trading .",
"we refer to such company - specific information as local information .",
"stock returns may also be influenced by global i... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "stock returns",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"stock",
"returns"
],
"offsets": [
28,
29
]
}
],
"trigger": {
... | [
"most",
"works",
"on",
"financial",
"forecasting",
"use",
"information",
"directly",
"associated",
"with",
"individual",
"companies",
"(",
"e",
".",
"g",
".",
",",
"stock",
"prices",
",",
"news",
"on",
"the",
"company",
")",
"to",
"predict",
"stock",
"return... |
ACL | Region-dependent temperature scaling for certainty calibration and application to class-imbalanced token classification | Certainty calibration is an important goal on the path to interpretability and trustworthy AI. Particularly in the context of human-in-the-loop systems, high-quality low to mid-range certainty estimates are essential. In the presence of a dominant high-certainty class, for instance the non-entity class in NER problems,... | 1e45b1eb873b27514d292318df1d6072 | 2,022 | [
"certainty calibration is an important goal on the path to interpretability and trustworthy ai .",
"particularly in the context of human - in - the - loop systems , high - quality low to mid - range certainty estimates are essential .",
"in the presence of a dominant high - certainty class , for instance the no... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "certainty calibration",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"certainty",
"calibration"
],
"offsets": [
0,
1
]
}
],
"tr... | [
"certainty",
"calibration",
"is",
"an",
"important",
"goal",
"on",
"the",
"path",
"to",
"interpretability",
"and",
"trustworthy",
"ai",
".",
"particularly",
"in",
"the",
"context",
"of",
"human",
"-",
"in",
"-",
"the",
"-",
"loop",
"systems",
",",
"high",
... |
ACL | Modeling Dual Read/Write Paths for Simultaneous Machine Translation | Simultaneous machine translation (SiMT) outputs translation while reading source sentence and hence requires a policy to decide whether to wait for the next source word (READ) or generate a target word (WRITE), the actions of which form a read/write path. Although the read/write path is essential to SiMT performance, n... | 055c111b6beca484f17a44df04c8734f | 2,022 | [
"simultaneous machine translation ( simt ) outputs translation while reading source sentence and hence requires a policy to decide whether to wait for the next source word ( read ) or generate a target word ( write ) , the actions of which form a read / write path .",
"although the read / write path is essential ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "simt",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"simt"
],
"offsets": [
87
]
}
],
"trigger": {
"text": "outputs",
"tokens": [
... | [
"simultaneous",
"machine",
"translation",
"(",
"simt",
")",
"outputs",
"translation",
"while",
"reading",
"source",
"sentence",
"and",
"hence",
"requires",
"a",
"policy",
"to",
"decide",
"whether",
"to",
"wait",
"for",
"the",
"next",
"source",
"word",
"(",
"re... |
ACL | Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision | The effectiveness of Neural Information Retrieval (Neu-IR) often depends on a large scale of in-domain relevance training signals, which are not always available in real-world ranking scenarios. To democratize the benefits of Neu-IR, this paper presents MetaAdaptRank, a domain adaptive learning method that generalizes ... | d6cb3c5479a6b731bec0782bd6ca70ec | 2,021 | [
"the effectiveness of neural information retrieval ( neu - ir ) often depends on a large scale of in - domain relevance training signals , which are not always available in real - world ranking scenarios .",
"to democratize the benefits of neu - ir , this paper presents metaadaptrank , a domain adaptive learning ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neu - ir",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neu",
"-",
"ir"
],
"offsets": [
42,
43,
44
]
}
],
... | [
"the",
"effectiveness",
"of",
"neural",
"information",
"retrieval",
"(",
"neu",
"-",
"ir",
")",
"often",
"depends",
"on",
"a",
"large",
"scale",
"of",
"in",
"-",
"domain",
"relevance",
"training",
"signals",
",",
"which",
"are",
"not",
"always",
"available",... |
ACL | Obtaining Better Static Word Embeddings Using Contextual Embedding Models | The advent of contextual word embeddings — representations of words which incorporate semantic and syntactic information from their context—has led to tremendous improvements on a wide variety of NLP tasks. However, recent contextual models have prohibitively high computational cost in many use-cases and are often hard... | 96a6dd1cc7474662ca1e1736444a75a3 | 2,021 | [
"the advent of contextual word embeddings — representations of words which incorporate semantic and syntactic information from their context — has led to tremendous improvements on a wide variety of nlp tasks .",
"however , recent contextual models have prohibitively high computational cost in many use - cases an... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "contextual word embeddings",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"contextual",
"word",
"embeddings"
],
"offsets": [
3,
4,
... | [
"the",
"advent",
"of",
"contextual",
"word",
"embeddings",
"—",
"representations",
"of",
"words",
"which",
"incorporate",
"semantic",
"and",
"syntactic",
"information",
"from",
"their",
"context",
"—",
"has",
"led",
"to",
"tremendous",
"improvements",
"on",
"a",
... |
ACL | Modeling Label Semantics for Predicting Emotional Reactions | Predicting how events induce emotions in the characters of a story is typically seen as a standard multi-label classification task, which usually treats labels as anonymous classes to predict. They ignore information that may be conveyed by the emotion labels themselves. We propose that the semantics of emotion labels ... | 5fc432f77498acdd4ef7f46ba5382c61 | 2,020 | [
"predicting how events induce emotions in the characters of a story is typically seen as a standard multi - label classification task , which usually treats labels as anonymous classes to predict .",
"they ignore information that may be conveyed by the emotion labels themselves .",
"we propose that the semantic... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "standard multi - label classification task",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"standard",
"multi",
"-",
"label",
"classification",
"t... | [
"predicting",
"how",
"events",
"induce",
"emotions",
"in",
"the",
"characters",
"of",
"a",
"story",
"is",
"typically",
"seen",
"as",
"a",
"standard",
"multi",
"-",
"label",
"classification",
"task",
",",
"which",
"usually",
"treats",
"labels",
"as",
"anonymous... |
ACL | Data-to-text Generation with Entity Modeling | Recent approaches to data-to-text generation have shown great promise thanks to the use of large-scale datasets and the application of neural network architectures which are trained end-to-end. These models rely on representation learning to select content appropriately, structure it coherently, and verbalize it gramma... | 9403a3baa146e616182c108c092d2dcf | 2,019 | [
"recent approaches to data - to - text generation have shown great promise thanks to the use of large - scale datasets and the application of neural network architectures which are trained end - to - end .",
"these models rely on representation learning to select content appropriately , structure it coherently , ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "approaches to data - to - text generation",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"approaches",
"to",
"data",
"-",
"to",
"-",
"t... | [
"recent",
"approaches",
"to",
"data",
"-",
"to",
"-",
"text",
"generation",
"have",
"shown",
"great",
"promise",
"thanks",
"to",
"the",
"use",
"of",
"large",
"-",
"scale",
"datasets",
"and",
"the",
"application",
"of",
"neural",
"network",
"architectures",
"... |
ACL | Improved Natural Language Generation via Loss Truncation | Neural language models are usually trained to match the distributional properties of large-scale corpora by minimizing the log loss. While straightforward to optimize, this approach forces the model to reproduce all variations in the dataset, including noisy and invalid references (e.g., misannotations and hallucinated... | 6099054c92e2bb896ec3358ac9167430 | 2,020 | [
"neural language models are usually trained to match the distributional properties of large - scale corpora by minimizing the log loss .",
"while straightforward to optimize , this approach forces the model to reproduce all variations in the dataset , including noisy and invalid references ( e . g . , misannotati... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neural language models",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neural",
"language",
"models"
],
"offsets": [
0,
1,
2
... | [
"neural",
"language",
"models",
"are",
"usually",
"trained",
"to",
"match",
"the",
"distributional",
"properties",
"of",
"large",
"-",
"scale",
"corpora",
"by",
"minimizing",
"the",
"log",
"loss",
".",
"while",
"straightforward",
"to",
"optimize",
",",
"this",
... |
ACL | CluBERT: A Cluster-Based Approach for Learning Sense Distributions in Multiple Languages | Knowing the Most Frequent Sense (MFS) of a word has been proved to help Word Sense Disambiguation (WSD) models significantly. However, the scarcity of sense-annotated data makes it difficult to induce a reliable and high-coverage distribution of the meanings in a language vocabulary. To address this issue, in this pape... | b58c20ab555455685d6b981bae947bd2 | 2,020 | [
"knowing the most frequent sense ( mfs ) of a word has been proved to help word sense disambiguation ( wsd ) models significantly .",
"however , the scarcity of sense - annotated data makes it difficult to induce a reliable and high - coverage distribution of the meanings in a language vocabulary .",
"to addres... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "word sense disambiguation models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"word",
"sense",
"disambiguation",
"models"
],
"offsets": [
... | [
"knowing",
"the",
"most",
"frequent",
"sense",
"(",
"mfs",
")",
"of",
"a",
"word",
"has",
"been",
"proved",
"to",
"help",
"word",
"sense",
"disambiguation",
"(",
"wsd",
")",
"models",
"significantly",
".",
"however",
",",
"the",
"scarcity",
"of",
"sense",
... |
ACL | Soft Representation Learning for Sparse Transfer | Transfer learning is effective for improving the performance of tasks that are related, and Multi-task learning (MTL) and Cross-lingual learning (CLL) are important instances. This paper argues that hard-parameter sharing, of hard-coding layers shared across different tasks or languages, cannot generalize well, when sh... | 855c019e24b2ca9d58d53779fe5a0e4d | 2,019 | [
"transfer learning is effective for improving the performance of tasks that are related , and multi - task learning ( mtl ) and cross - lingual learning ( cll ) are important instances .",
"this paper argues that hard - parameter sharing , of hard - coding layers shared across different tasks or languages , canno... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "transfer learning",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"transfer",
"learning"
],
"offsets": [
0,
1
]
}
],
"trigger": ... | [
"transfer",
"learning",
"is",
"effective",
"for",
"improving",
"the",
"performance",
"of",
"tasks",
"that",
"are",
"related",
",",
"and",
"multi",
"-",
"task",
"learning",
"(",
"mtl",
")",
"and",
"cross",
"-",
"lingual",
"learning",
"(",
"cll",
")",
"are",... |
ACL | Fine-Grained Controllable Text Generation Using Non-Residual Prompting | The introduction of immensely large Causal Language Models (CLMs) has rejuvenated the interest in open-ended text generation. However, controlling the generative process for these Transformer-based models is at large an unsolved problem. Earlier work has explored either plug-and-play decoding strategies, or more powerf... | 59330a3cd9a15952f51f8be3b202d292 | 2,022 | [
"the introduction of immensely large causal language models ( clms ) has rejuvenated the interest in open - ended text generation .",
"however , controlling the generative process for these transformer - based models is at large an unsolved problem .",
"earlier work has explored either plug - and - play decodin... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "immensely large causal language models",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"immensely",
"large",
"causal",
"language",
"models"
],
... | [
"the",
"introduction",
"of",
"immensely",
"large",
"causal",
"language",
"models",
"(",
"clms",
")",
"has",
"rejuvenated",
"the",
"interest",
"in",
"open",
"-",
"ended",
"text",
"generation",
".",
"however",
",",
"controlling",
"the",
"generative",
"process",
... |
ACL | Things not Written in Text: Exploring Spatial Commonsense from Visual Signals | Spatial commonsense, the knowledge about spatial position and relationship between objects (like the relative size of a lion and a girl, and the position of a boy relative to a bicycle when cycling), is an important part of commonsense knowledge. Although pretrained language models (PLMs) succeed in many NLP tasks, the... | 50c9dc936b554c4e17f08358c89f1ce0 | 2,022 | [
"spatial commonsense , the knowledge about spatial position and relationship between objects ( like the relative size of a lion and a girl , and the position of a boy relative to a bicycle when cycling ) , is an important part of commonsense knowledge .",
"although pretrained language models ( plms ) succeed in m... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "commonsense knowledge",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"commonsense",
"knowledge"
],
"offsets": [
43,
44
]
}
],
"... | [
"spatial",
"commonsense",
",",
"the",
"knowledge",
"about",
"spatial",
"position",
"and",
"relationship",
"between",
"objects",
"(",
"like",
"the",
"relative",
"size",
"of",
"a",
"lion",
"and",
"a",
"girl",
",",
"and",
"the",
"position",
"of",
"a",
"boy",
... |
ACL | Uncertainty-Aware Curriculum Learning for Neural Machine Translation | Neural machine translation (NMT) has proven to be facilitated by curriculum learning which presents examples in an easy-to-hard order at different training stages. The keys lie in the assessment of data difficulty and model competence. We propose uncertainty-aware curriculum learning, which is motivated by the intuitio... | d05296dc3679d6ea318032ac4563aeae | 2,020 | [
"neural machine translation ( nmt ) has proven to be facilitated by curriculum learning which presents examples in an easy - to - hard order at different training stages .",
"the keys lie in the assessment of data difficulty and model competence .",
"we propose uncertainty - aware curriculum learning , which is... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "neural machine translation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"neural",
"machine",
"translation"
],
"offsets": [
0,
1,
... | [
"neural",
"machine",
"translation",
"(",
"nmt",
")",
"has",
"proven",
"to",
"be",
"facilitated",
"by",
"curriculum",
"learning",
"which",
"presents",
"examples",
"in",
"an",
"easy",
"-",
"to",
"-",
"hard",
"order",
"at",
"different",
"training",
"stages",
".... |
ACL | Effective Token Graph Modeling using a Novel Labeling Strategy for Structured Sentiment Analysis | The state-of-the-art model for structured sentiment analysis casts the task as a dependency parsing problem, which has some limitations: (1) The label proportions for span prediction and span relation prediction are imbalanced. (2) The span lengths of sentiment tuple components may be very large in this task, which wil... | 4462428c23da73b8c20734f5d40aa19b | 2,022 | [
"the state - of - the - art model for structured sentiment analysis casts the task as a dependency parsing problem , which has some limitations : ( 1 ) the label proportions for span prediction and span relation prediction are imbalanced .",
"( 2 ) the span lengths of sentiment tuple components may be very large ... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "structured sentiment analysis",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"structured",
"sentiment",
"analysis"
],
"offsets": [
10,
11,
... | [
"the",
"state",
"-",
"of",
"-",
"the",
"-",
"art",
"model",
"for",
"structured",
"sentiment",
"analysis",
"casts",
"the",
"task",
"as",
"a",
"dependency",
"parsing",
"problem",
",",
"which",
"has",
"some",
"limitations",
":",
"(",
"1",
")",
"the",
"label... |
ACL | Diachronic Sense Modeling with Deep Contextualized Word Embeddings: An Ecological View | Diachronic word embeddings have been widely used in detecting temporal changes. However, existing methods face the meaning conflation deficiency by representing a word as a single vector at each time period. To address this issue, this paper proposes a sense representation and tracking framework based on deep contextua... | 528a8bccaccbdcacb46e6956a07f43e8 | 2,019 | [
"diachronic word embeddings have been widely used in detecting temporal changes .",
"however , existing methods face the meaning conflation deficiency by representing a word as a single vector at each time period .",
"to address this issue , this paper proposes a sense representation and tracking framework base... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "diachronic word embeddings",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"diachronic",
"word",
"embeddings"
],
"offsets": [
0,
1,
... | [
"diachronic",
"word",
"embeddings",
"have",
"been",
"widely",
"used",
"in",
"detecting",
"temporal",
"changes",
".",
"however",
",",
"existing",
"methods",
"face",
"the",
"meaning",
"conflation",
"deficiency",
"by",
"representing",
"a",
"word",
"as",
"a",
"singl... |
ACL | Nested Named Entity Recognition via Explicitly Excluding the Influence of the Best Path | This paper presents a novel method for nested named entity recognition. As a layered method, our method extends the prior second-best path recognition method by explicitly excluding the influence of the best path. Our method maintains a set of hidden states at each time step and selectively leverages them to build a di... | f083db2d650b7acf9eabb2d9f79585e5 | 2,021 | [
"this paper presents a novel method for nested named entity recognition .",
"as a layered method , our method extends the prior second - best path recognition method by explicitly excluding the influence of the best path .",
"our method maintains a set of hidden states at each time step and selectively leverage... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "method",
"nugget_type": "APP",
"argument_type": "Content",
"tokens": [
"method"
],
"offsets": [
5
]
},
{
"text": "nested named entity recognition",
"... | [
"this",
"paper",
"presents",
"a",
"novel",
"method",
"for",
"nested",
"named",
"entity",
"recognition",
".",
"as",
"a",
"layered",
"method",
",",
"our",
"method",
"extends",
"the",
"prior",
"second",
"-",
"best",
"path",
"recognition",
"method",
"by",
"expli... |
ACL | Detecting Unassimilated Borrowings in Spanish: An Annotated Corpus and Approaches to Modeling | This work presents a new resource for borrowing identification and analyzes the performance and errors of several models on this task. We introduce a new annotated corpus of Spanish newswire rich in unassimilated lexical borrowings—words from one language that are introduced into another without orthographic adaptation... | 5b8511803616a0bef1f80e6a1677bdec | 2,022 | [
"this work presents a new resource for borrowing identification and analyzes the performance and errors of several models on this task .",
"we introduce a new annotated corpus of spanish newswire rich in unassimilated lexical borrowings — words from one language that are introduced into another without orthograph... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
22
]
},
{
"text": "annotated corpus of spanish newswire rich",
... | [
"this",
"work",
"presents",
"a",
"new",
"resource",
"for",
"borrowing",
"identification",
"and",
"analyzes",
"the",
"performance",
"and",
"errors",
"of",
"several",
"models",
"on",
"this",
"task",
".",
"we",
"introduce",
"a",
"new",
"annotated",
"corpus",
"of"... |
ACL | BERT Learns to Teach: Knowledge Distillation with Meta Learning | We present Knowledge Distillation with Meta Learning (MetaDistil), a simple yet effective alternative to traditional knowledge distillation (KD) methods where the teacher model is fixed during training. We show the teacher network can learn to better transfer knowledge to the student network (i.e., learning to teach) w... | beb8197b6da785aa075e6116eb74fd55 | 2,022 | [
"we present knowledge distillation with meta learning ( metadistil ) , a simple yet effective alternative to traditional knowledge distillation ( kd ) methods where the teacher model is fixed during training .",
"we show the teacher network can learn to better transfer knowledge to the student network ( i . e . ,... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "knowledge distillation with meta learning",
... | [
"we",
"present",
"knowledge",
"distillation",
"with",
"meta",
"learning",
"(",
"metadistil",
")",
",",
"a",
"simple",
"yet",
"effective",
"alternative",
"to",
"traditional",
"knowledge",
"distillation",
"(",
"kd",
")",
"methods",
"where",
"the",
"teacher",
"mode... |
ACL | Bias Mitigation in Machine Translation Quality Estimation | Machine Translation Quality Estimation (QE) aims to build predictive models to assess the quality of machine-generated translations in the absence of reference translations. While state-of-the-art QE models have been shown to achieve good results, they over-rely on features that do not have a causal impact on the quali... | 52f73a7b9fb1fefb33f9e9af3e28e3e4 | 2,022 | [
"machine translation quality estimation ( qe ) aims to build predictive models to assess the quality of machine - generated translations in the absence of reference translations .",
"while state - of - the - art qe models have been shown to achieve good results , they over - rely on features that do not have a ca... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "machine translation quality estimation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"machine",
"translation",
"quality",
"estimation"
],
"offsets":... | [
"machine",
"translation",
"quality",
"estimation",
"(",
"qe",
")",
"aims",
"to",
"build",
"predictive",
"models",
"to",
"assess",
"the",
"quality",
"of",
"machine",
"-",
"generated",
"translations",
"in",
"the",
"absence",
"of",
"reference",
"translations",
".",... |
ACL | StereoSet: Measuring stereotypical bias in pretrained language models | A stereotype is an over-generalized belief about a particular group of people, e.g., Asians are good at math or African Americans are athletic. Such beliefs (biases) are known to hurt target groups. Since pretrained language models are trained on large real-world data, they are known to capture stereotypical biases. It... | 5944d93829d9bc0dd9de692165862f1f | 2,021 | [
"a stereotype is an over - generalized belief about a particular group of people , e . g . , asians are good at math or african americans are athletic .",
"such beliefs ( biases ) are known to hurt target groups .",
"since pretrained language models are trained on large real - world data , they are known to cap... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "stereotype",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"stereotype"
],
"offsets": [
1
]
}
],
"trigger": {
"text": "belief",
"tokens"... | [
"a",
"stereotype",
"is",
"an",
"over",
"-",
"generalized",
"belief",
"about",
"a",
"particular",
"group",
"of",
"people",
",",
"e",
".",
"g",
".",
",",
"asians",
"are",
"good",
"at",
"math",
"or",
"african",
"americans",
"are",
"athletic",
".",
"such",
... |
ACL | Conditional Bilingual Mutual Information Based Adaptive Training for Neural Machine Translation | Token-level adaptive training approaches can alleviate the token imbalance problem and thus improve neural machine translation, through re-weighting the losses of different target tokens based on specific statistical metrics (e.g., token frequency or mutual information). Given that standard translation models make pred... | 908c85fe8bb43e61635b58b64c14b6ba | 2,022 | [
"token - level adaptive training approaches can alleviate the token imbalance problem and thus improve neural machine translation , through re - weighting the losses of different target tokens based on specific statistical metrics ( e . g . , token frequency or mutual information ) .",
"given that standard transl... | [
{
"event_type": "RWS",
"arguments": [
{
"text": "token - level adaptive training approaches",
"nugget_type": "APP",
"argument_type": "Subject",
"tokens": [
"token",
"-",
"level",
"adaptive",
"training",
"approa... | [
"token",
"-",
"level",
"adaptive",
"training",
"approaches",
"can",
"alleviate",
"the",
"token",
"imbalance",
"problem",
"and",
"thus",
"improve",
"neural",
"machine",
"translation",
",",
"through",
"re",
"-",
"weighting",
"the",
"losses",
"of",
"different",
"ta... |
ACL | On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation | Adapter-based tuning has recently arisen as an alternative to fine-tuning. It works by adding light-weight adapter modules to a pretrained language model (PrLM) and only updating the parameters of adapter modules when learning on a downstream task. As such, it adds only a few trainable parameters per new task, allowing... | a9853d5805c5355a564165d5045e0987 | 2,021 | [
"adapter - based tuning has recently arisen as an alternative to fine - tuning .",
"it works by adding light - weight adapter modules to a pretrained language model ( prlm ) and only updating the parameters of adapter modules when learning on a downstream task .",
"as such , it adds only a few trainable paramet... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "adapter - based tuning",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"adapter",
"-",
"based",
"tuning"
],
"offsets": [
0,
1,
... | [
"adapter",
"-",
"based",
"tuning",
"has",
"recently",
"arisen",
"as",
"an",
"alternative",
"to",
"fine",
"-",
"tuning",
".",
"it",
"works",
"by",
"adding",
"light",
"-",
"weight",
"adapter",
"modules",
"to",
"a",
"pretrained",
"language",
"model",
"(",
"pr... |
ACL | Adversarial Domain Adaptation Using Artificial Titles for Abstractive Title Generation | A common issue in training a deep learning, abstractive summarization model is lack of a large set of training summaries. This paper examines techniques for adapting from a labeled source domain to an unlabeled target domain in the context of an encoder-decoder model for text generation. In addition to adversarial doma... | 1396f792b2ffcad939df10402533d724 | 2,019 | [
"a common issue in training a deep learning , abstractive summarization model is lack of a large set of training summaries .",
"this paper examines techniques for adapting from a labeled source domain to an unlabeled target domain in the context of an encoder - decoder model for text generation .",
"in addition... | [
{
"event_type": "RWF",
"arguments": [],
"trigger": {
"text": "lack",
"tokens": [
"lack"
],
"offsets": [
13
]
}
},
{
"event_type": "WKS",
"arguments": [
{
"text": "techniques for adapting from a labeled source domain to an unlabe... | [
"a",
"common",
"issue",
"in",
"training",
"a",
"deep",
"learning",
",",
"abstractive",
"summarization",
"model",
"is",
"lack",
"of",
"a",
"large",
"set",
"of",
"training",
"summaries",
".",
"this",
"paper",
"examines",
"techniques",
"for",
"adapting",
"from",
... |
ACL | Informative Image Captioning with External Sources of Information | An image caption should fluently present the essential information in a given image, including informative, fine-grained entity mentions and the manner in which these entities interact. However, current captioning models are usually trained to generate captions that only contain common object names, thus falling short ... | 80feef04b28a70470cf558717375ec20 | 2,019 | [
"an image caption should fluently present the essential information in a given image , including informative , fine - grained entity mentions and the manner in which these entities interact .",
"however , current captioning models are usually trained to generate captions that only contain common object names , th... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "image caption",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"image",
"caption"
],
"offsets": [
1,
2
]
}
],
"trigger": {
... | [
"an",
"image",
"caption",
"should",
"fluently",
"present",
"the",
"essential",
"information",
"in",
"a",
"given",
"image",
",",
"including",
"informative",
",",
"fine",
"-",
"grained",
"entity",
"mentions",
"and",
"the",
"manner",
"in",
"which",
"these",
"enti... |
ACL | UniXcoder: Unified Cross-Modal Pre-training for Code Representation | Pre-trained models for programming languages have recently demonstrated great success on code intelligence. To support both code-related understanding and generation tasks, recent works attempt to pre-train unified encoder-decoder models. However, such encoder-decoder framework is sub-optimal for auto-regressive tasks,... | 6c02d8681be0152f7f9ae9829743f365 | 2,022 | [
"pre - trained models for programming languages have recently demonstrated great success on code intelligence .",
"to support both code - related understanding and generation tasks , recent works attempt to pre - train unified encoder - decoder models .",
"however , such encoder - decoder framework is sub - opt... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "programming languages",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"programming",
"languages"
],
"offsets": [
5,
6
]
}
],
"tr... | [
"pre",
"-",
"trained",
"models",
"for",
"programming",
"languages",
"have",
"recently",
"demonstrated",
"great",
"success",
"on",
"code",
"intelligence",
".",
"to",
"support",
"both",
"code",
"-",
"related",
"understanding",
"and",
"generation",
"tasks",
",",
"r... |
ACL | A Just and Comprehensive Strategy for Using NLP to Address Online Abuse | Online abusive behavior affects millions and the NLP community has attempted to mitigate this problem by developing technologies to detect abuse. However, current methods have largely focused on a narrow definition of abuse to detriment of victims who seek both validation and solutions. In this position paper, we argue... | 32eac87611733f9b146d0fec8732682b | 2,019 | [
"online abusive behavior affects millions and the nlp community has attempted to mitigate this problem by developing technologies to detect abuse .",
"however , current methods have largely focused on a narrow definition of abuse to detriment of victims who seek both validation and solutions .",
"in this positi... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "abuse",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"abuse"
],
"offsets": [
20
]
}
],
"trigger": {
"text": "detect",
"tokens": [
... | [
"online",
"abusive",
"behavior",
"affects",
"millions",
"and",
"the",
"nlp",
"community",
"has",
"attempted",
"to",
"mitigate",
"this",
"problem",
"by",
"developing",
"technologies",
"to",
"detect",
"abuse",
".",
"however",
",",
"current",
"methods",
"have",
"la... |
ACL | A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction | Neural models have achieved great success on machine reading comprehension (MRC), many of which typically consist of two components: an evidence extractor and an answer predictor. The former seeks the most relevant information from a reference text, while the latter is to locate or generate answers from the extracted e... | fa65b2c93afe0f7fbb681c6fed42380b | 2,020 | [
"neural models have achieved great success on machine reading comprehension ( mrc ) , many of which typically consist of two components : an evidence extractor and an answer predictor .",
"the former seeks the most relevant information from a reference text , while the latter is to locate or generate answers from... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "mrc",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"mrc"
],
"offsets": [
134
]
}
],
"trigger": {
"text": "achieved",
"tokens": [
... | [
"neural",
"models",
"have",
"achieved",
"great",
"success",
"on",
"machine",
"reading",
"comprehension",
"(",
"mrc",
")",
",",
"many",
"of",
"which",
"typically",
"consist",
"of",
"two",
"components",
":",
"an",
"evidence",
"extractor",
"and",
"an",
"answer",
... |
ACL | Capturing Relations between Scientific Papers: An Abstractive Model for Related Work Section Generation | Given a set of related publications, related work section generation aims to provide researchers with an overview of the specific research area by summarizing these works and introducing them in a logical order. Most of existing related work generation models follow the inflexible extractive style, which directly extra... | b1cc4941ec3a3f59fcdca496be1618d7 | 2,021 | [
"given a set of related publications , related work section generation aims to provide researchers with an overview of the specific research area by summarizing these works and introducing them in a logical order .",
"most of existing related work generation models follow the inflexible extractive style , which d... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "related work section generation",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"related",
"work",
"section",
"generation"
],
"offsets": [
7... | [
"given",
"a",
"set",
"of",
"related",
"publications",
",",
"related",
"work",
"section",
"generation",
"aims",
"to",
"provide",
"researchers",
"with",
"an",
"overview",
"of",
"the",
"specific",
"research",
"area",
"by",
"summarizing",
"these",
"works",
"and",
... |
ACL | Entity Enhancement for Implicit Discourse Relation Classification in the Biomedical Domain | Implicit discourse relation classification is a challenging task, in particular when the text domain is different from the standard Penn Discourse Treebank (PDTB; Prasad et al., 2008) training corpus domain (Wall Street Journal in 1990s). We here tackle the task of implicit discourse relation classification on the biom... | 92e16cb75628d173370ff4317abd286c | 2,021 | [
"implicit discourse relation classification is a challenging task , in particular when the text domain is different from the standard penn discourse treebank ( pdtb ; prasad et al . , 2008 ) training corpus domain ( wall street journal in 1990s ) .",
"we here tackle the task of implicit discourse relation classif... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "implicit discourse relation classification",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"implicit",
"discourse",
"relation",
"classification"
],
"o... | [
"implicit",
"discourse",
"relation",
"classification",
"is",
"a",
"challenging",
"task",
",",
"in",
"particular",
"when",
"the",
"text",
"domain",
"is",
"different",
"from",
"the",
"standard",
"penn",
"discourse",
"treebank",
"(",
"pdtb",
";",
"prasad",
"et",
... |
ACL | Harvesting and Refining Question-Answer Pairs for Unsupervised QA | Question Answering (QA) has shown great success thanks to the availability of large-scale datasets and the effectiveness of neural models. Recent research works have attempted to extend these successes to the settings with few or no labeled data available. In this work, we introduce two approaches to improve unsupervis... | ba8b50b0a981b2337d2d549ef46e7ef6 | 2,020 | [
"question answering ( qa ) has shown great success thanks to the availability of large - scale datasets and the effectiveness of neural models .",
"recent research works have attempted to extend these successes to the settings with few or no labeled data available .",
"in this work , we introduce two approaches... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "question answering",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"question",
"answering"
],
"offsets": [
0,
1
]
}
],
"trigger"... | [
"question",
"answering",
"(",
"qa",
")",
"has",
"shown",
"great",
"success",
"thanks",
"to",
"the",
"availability",
"of",
"large",
"-",
"scale",
"datasets",
"and",
"the",
"effectiveness",
"of",
"neural",
"models",
".",
"recent",
"research",
"works",
"have",
... |
ACL | “You Sound Just Like Your Father” Commercial Machine Translation Systems Include Stylistic Biases | The main goal of machine translation has been to convey the correct content. Stylistic considerations have been at best secondary. We show that as a consequence, the output of three commercial machine translation systems (Bing, DeepL, Google) make demographically diverse samples from five languages “sound” older and mo... | 77ebbaf5ac1becbc58fe18e16e13a896 | 2,020 | [
"the main goal of machine translation has been to convey the correct content .",
"stylistic considerations have been at best secondary .",
"we show that as a consequence , the output of three commercial machine translation systems ( bing , deepl , google ) make demographically diverse samples from five language... | [
{
"event_type": "FIN",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Finder",
"tokens": [
"we"
],
"offsets": [
22
]
},
{
"text": "sound older more male",
"nugget_type": "E-CM... | [
"the",
"main",
"goal",
"of",
"machine",
"translation",
"has",
"been",
"to",
"convey",
"the",
"correct",
"content",
".",
"stylistic",
"considerations",
"have",
"been",
"at",
"best",
"secondary",
".",
"we",
"show",
"that",
"as",
"a",
"consequence",
",",
"the",... |
ACL | Morphological Irregularity Correlates with Frequency | We present a study of morphological irregularity. Following recent work, we define an information-theoretic measure of irregularity based on the predictability of forms in a language. Using a neural transduction model, we estimate this quantity for the forms in 28 languages. We first present several validatory and expl... | 6080feecdcd3579f3d85946db4ab58b4 | 2,019 | [
"we present a study of morphological irregularity .",
"following recent work , we define an information - theoretic measure of irregularity based on the predictability of forms in a language .",
"using a neural transduction model , we estimate this quantity for the forms in 28 languages .",
"we first present ... | [
{
"event_type": "PRP",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Proposer",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "study of morphological irregularity",
"nugg... | [
"we",
"present",
"a",
"study",
"of",
"morphological",
"irregularity",
".",
"following",
"recent",
"work",
",",
"we",
"define",
"an",
"information",
"-",
"theoretic",
"measure",
"of",
"irregularity",
"based",
"on",
"the",
"predictability",
"of",
"forms",
"in",
... |
ACL | On the Robustness of Language Encoders against Grammatical Errors | We conduct a thorough study to diagnose the behaviors of pre-trained language encoders (ELMo, BERT, and RoBERTa) when confronted with natural grammatical errors. Specifically, we collect real grammatical errors from non-native speakers and conduct adversarial attacks to simulate these errors on clean text data. We use ... | 86af4b6f6104bd2a105d026057f7fa21 | 2,020 | [
"we conduct a thorough study to diagnose the behaviors of pre - trained language encoders ( elmo , bert , and roberta ) when confronted with natural grammatical errors .",
"specifically , we collect real grammatical errors from non - native speakers and conduct adversarial attacks to simulate these errors on clea... | [
{
"event_type": "WKS",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Researcher",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "thorough study",
"nugget_type": "APP",
... | [
"we",
"conduct",
"a",
"thorough",
"study",
"to",
"diagnose",
"the",
"behaviors",
"of",
"pre",
"-",
"trained",
"language",
"encoders",
"(",
"elmo",
",",
"bert",
",",
"and",
"roberta",
")",
"when",
"confronted",
"with",
"natural",
"grammatical",
"errors",
".",... |
ACL | Reservoir Transformers | We demonstrate that transformers obtain impressive performance even when some of the layers are randomly initialized and never updated. Inspired by old and well-established ideas in machine learning, we explore a variety of non-linear “reservoir” layers interspersed with regular transformer layers, and show improvement... | 75501e50b64ae30a5b8fcd1168e39efd | 2,021 | [
"we demonstrate that transformers obtain impressive performance even when some of the layers are randomly initialized and never updated .",
"inspired by old and well - established ideas in machine learning , we explore a variety of non - linear “ reservoir ” layers interspersed with regular transformer layers , a... | [
{
"event_type": "FIN",
"arguments": [
{
"text": "we",
"nugget_type": "OG",
"argument_type": "Finder",
"tokens": [
"we"
],
"offsets": [
0
]
},
{
"text": "obtain",
"nugget_type": "E-FAC",
"arg... | [
"we",
"demonstrate",
"that",
"transformers",
"obtain",
"impressive",
"performance",
"even",
"when",
"some",
"of",
"the",
"layers",
"are",
"randomly",
"initialized",
"and",
"never",
"updated",
".",
"inspired",
"by",
"old",
"and",
"well",
"-",
"established",
"idea... |
ACL | BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation | Intrinsic evaluations of OIE systems are carried out either manually—with human evaluators judging the correctness of extractions—or automatically, on standardized benchmarks. The latter, while much more cost-effective, is less reliable, primarily because of the incompleteness of the existing OIE benchmarks: the ground... | cf4a05fe39f40fe6dd4f5c9513292124 | 2,022 | [
"intrinsic evaluations of oie systems are carried out either manually — with human evaluators judging the correctness of extractions — or automatically , on standardized benchmarks .",
"the latter , while much more cost - effective , is less reliable , primarily because of the incompleteness of the existing oie b... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "oie systems",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"oie",
"systems"
],
"offsets": [
3,
4
]
}
],
"trigger": {
"tex... | [
"intrinsic",
"evaluations",
"of",
"oie",
"systems",
"are",
"carried",
"out",
"either",
"manually",
"—",
"with",
"human",
"evaluators",
"judging",
"the",
"correctness",
"of",
"extractions",
"—",
"or",
"automatically",
",",
"on",
"standardized",
"benchmarks",
".",
... |
ACL | Predicting Difficulty and Discrimination of Natural Language Questions | Item Response Theory (IRT) has been extensively used to numerically characterize question difficulty and discrimination for human subjects in domains including cognitive psychology and education (Primi et al., 2014; Downing, 2003). More recently, IRT has been used to similarly characterize item difficulty and discrimin... | 7c77c44123b1f3eca780d4aac91b23f5 | 2,022 | [
"item response theory ( irt ) has been extensively used to numerically characterize question difficulty and discrimination for human subjects in domains including cognitive psychology and education ( primi et al . , 2014 ; downing , 2003 ) .",
"more recently , irt has been used to similarly characterize item diff... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "item response theory",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"item",
"response",
"theory"
],
"offsets": [
0,
1,
2
... | [
"item",
"response",
"theory",
"(",
"irt",
")",
"has",
"been",
"extensively",
"used",
"to",
"numerically",
"characterize",
"question",
"difficulty",
"and",
"discrimination",
"for",
"human",
"subjects",
"in",
"domains",
"including",
"cognitive",
"psychology",
"and",
... |
ACL | Bag-of-Words vs. Graph vs. Sequence in Text Classification: Questioning the Necessity of Text-Graphs and the Surprising Strength of a Wide MLP | Graph neural networks have triggered a resurgence of graph-based text classification methods, defining today’s state of the art. We show that a wide multi-layer perceptron (MLP) using a Bag-of-Words (BoW) outperforms the recent graph-based models TextGCN and HeteGCN in an inductive text classification setting and is co... | 335c996c88833436fbdcb784f5c23eaa | 2,022 | [
"graph neural networks have triggered a resurgence of graph - based text classification methods , defining today ’ s state of the art .",
"we show that a wide multi - layer perceptron ( mlp ) using a bag - of - words ( bow ) outperforms the recent graph - based models textgcn and hetegcn in an inductive text clas... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "graph neural networks",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"graph",
"neural",
"networks"
],
"offsets": [
0,
1,
2
... | [
"graph",
"neural",
"networks",
"have",
"triggered",
"a",
"resurgence",
"of",
"graph",
"-",
"based",
"text",
"classification",
"methods",
",",
"defining",
"today",
"’",
"s",
"state",
"of",
"the",
"art",
".",
"we",
"show",
"that",
"a",
"wide",
"multi",
"-",
... |
ACL | SMedBERT: A Knowledge-Enhanced Pre-trained Language Model with Structured Semantics for Medical Text Mining | Recently, the performance of Pre-trained Language Models (PLMs) has been significantly improved by injecting knowledge facts to enhance their abilities of language understanding. For medical domains, the background knowledge sources are especially useful, due to the massive medical terms and their complicated relations... | 083b8a9c28f0820f55818f14edf2ff2c | 2,021 | [
"recently , the performance of pre - trained language models ( plms ) has been significantly improved by injecting knowledge facts to enhance their abilities of language understanding .",
"for medical domains , the background knowledge sources are especially useful , due to the massive medical terms and their com... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "pre - trained language models",
"nugget_type": "APP",
"argument_type": "Target",
"tokens": [
"pre",
"-",
"trained",
"language",
"models"
],
"offsets": [
... | [
"recently",
",",
"the",
"performance",
"of",
"pre",
"-",
"trained",
"language",
"models",
"(",
"plms",
")",
"has",
"been",
"significantly",
"improved",
"by",
"injecting",
"knowledge",
"facts",
"to",
"enhance",
"their",
"abilities",
"of",
"language",
"understandi... |
ACL | GTM: A Generative Triple-wise Model for Conversational Question Generation | Generating some appealing questions in open-domain conversations is an effective way to improve human-machine interactions and lead the topic to a broader or deeper direction. To avoid dull or deviated questions, some researchers tried to utilize answer, the “future” information, to guide question generation. However, ... | 3ccd90a0f41c31d705045872a467a13e | 2,021 | [
"generating some appealing questions in open - domain conversations is an effective way to improve human - machine interactions and lead the topic to a broader or deeper direction .",
"to avoid dull or deviated questions , some researchers tried to utilize answer , the “ future ” information , to guide question g... | [
{
"event_type": "ITT",
"arguments": [
{
"text": "generating some appealing questions in open - domain conversations",
"nugget_type": "TAK",
"argument_type": "Target",
"tokens": [
"generating",
"some",
"appealing",
"questions",
... | [
"generating",
"some",
"appealing",
"questions",
"in",
"open",
"-",
"domain",
"conversations",
"is",
"an",
"effective",
"way",
"to",
"improve",
"human",
"-",
"machine",
"interactions",
"and",
"lead",
"the",
"topic",
"to",
"a",
"broader",
"or",
"deeper",
"direct... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.